Critical Care Obstetrics part 21 ppsx

Dialysis 189 The physiology of dialysis is based on diffusive and convective transport. Diffusion refers to the random movement of a solute down its concentration gradient. It is by this means that the majority of urea and solute clearance is achieved. Convection is that solute movement that occurs by means of solvent drag as water is removed, either by hydrostatic or osmotic force. A lesser degree of clearance is obtained during fl uid removal by ultrafi ltration. Modes of d ialysis Options for dialysis include hemodialysis and peritoneal dialysis, with the latter consisting of continuous ambulatory peritoneal dialysis (CAPD), continuous cycling peritoneal dialysis (CCPD), and nocturnal intermittent peritoneal dialysis (NIPD). Hemodialysis Hemodialysis requires a vascular access for extracorporeal therapy. This is usually a surgically created artifi cial arteriovenous (AV) shunt or a native AV fi stula, although dual - lumen central venous catheters can be used temporarily (Figure 13.1 ). Products of protein metabolism, such as urea nitrogen, potassium, and phosphate, are removed by both diffusion and convection across a semipermeable dialyzer membrane, while ions such as bicar- bonate and calcium diffuse into the blood. Fluid removal is accomplished by applying hydrostatic pressure across the dialyzer membrane. The dialysis prescription for non - pregnant patients generally consists of 3 – 4 hours of hemodialysis thrice weekly, depending on urea generation rate and dialyzer solute clearance. Heparinization is generally employed throughout the dialysis treatment. Peritoneal d ialysis The various forms of peritoneal dialysis have in common the removal of these same metabolites and excess fl uid, albeit by dif- fusion and convective fl ow across the peritoneal membrane. Surgical placement of a peritoneal catheter allows repeated access to the peritoneal cavity (Figure 13.2 ). Removal of fl uid by osmotic force is achieved by instilling a hypertonic dialysate such as dex- trose solution into the peritoneal cavity. Urea and other ions develop, as are fl uid overload and uremic complications (Table 13.2 ). In patients with diabetes who often have other end - organ damage, including autonomic neuropathy and vascular disease, dialytic support may be required even earlier, when the GFR reaches 15 mL/min. Table 13.1 Signs and symptoms of uremia. Organ involvement Subjective complaints Objective fi ndings Neurologic Cognitive diffi culties Hyperrefl exia, asterixis Sleep – wake reversal Seizures, encephalopathy Dysesthesias Peripheral neuropathy Hematopoietic Easy bruising and bleeding Anemia Fatigue Prolonged bleeding time Gastrointestinal Metallic taste Angiodysplasia Constipation Nausea Musculoskeletal Weakness Carpal tunnel syndrome Bone pain Bone fractures Myopathy Cardiovascular Dyspnea Hypertension Chest pain Pulmonary edema Pericarditis Dermatologic Pruritus Cutaneous calcifi cations Endocrine Decreased libido Decreased fertility Dysmenorrhea, amenorrhea Table 13.2 Indications for initiation of dialysis. Hyperkalemia Metabolic acidosis Volume overload Uremic pericarditis Uremic encephalopathy Glomerular fi ltration rate (GFR) 5 – 10 mL/min Figure 13.1 Hemodialysis. Chapter 13 190 cies in dialysis patients at their institution since 1965 [11] , although nearly one - third of their patients conceived before the onset of dialysis. There are many theoretical reasons to utilize peritoneal dialysis in pregnancy, most notably of which is the steady - state removal of uremic toxins (Table 13.3 ). This, coupled with easier fl uid removal, should minimize episodes of hypotension and thus pla- cental insuffi ciency. Additional advantages of peritoneal dialysis often include less severe anemia, as well as better blood pressure control and more liberal dietary restrictions due to the continu- ous nature of the therapy [12 – 14] . Furthermore, peritoneal dialy- sis obviates the need for systemic anticoagulation. In diabetic patients, the use of intraperitoneal insulin can also facilitate strict glycemic control. There have also been several case reports of successful intraperitoneal magnesium administration for the treatment of pre - eclampsia, maintaining a steady - state magne- sium serum level of approximately 5 mEq/L, although generally, alternative therapy may be recommended in renal failure to avoid magnesium toxicity [10,15] . Despite these apparent advantages of peritoneal dialysis, several unique complications exist, including catheter - related complica- tions such as laceration of the uterine vessels [16] and peritonitis. Hou reported precipitation of preterm labor and delivery in two of three patients secondary to peritonitis, but other reports suggest that the incidence of peritonitis is not increased in preg- nant versus non - pregnant patients [6] . Peritoneal dialysis cath- eters have been placed as late as 29 weeks gestation. In some patients, however, diffi culties with catheter obstruction and failure to drain necessitate placement of multiple catheters or conversion to hemodialysis. It is also diffi cult to determine whether either method of dialysis actually precipitates preterm labor, because preterm labor has been described in the setting of both hemodialysis and peritoneal dialysis, as well as in CRF alone. Intensive d ialysis Generally, modifi cation of the dialysis prescription in pregnancy has been recommended for patients treated with both hemodialy- present in high concentrations diffuse from the peritoneal vascu- lature into the dialysate, while calcium and a bicarbonate source such as lactate move in the opposite direction. Depending on the mode of peritoneal dialysis selected, dialysate is instilled and drained either manually or automatically at repeated intervals throughout the day. CAPD consists of approximately four manual exchanges per day; the peritoneum is fi lled with several liters of dialysate with each exchange, and the fl uid is drained 4 – 6 hours later. Both CCPD and NIPD utilize an automated cycler to repeatedly fi ll and drain the peritoneum at shorter intervals throughout the night. CCPD differs in that it also includes a daytime dwell for added clearance. Dialysis and p regnancy Hemodialysis v s p eritoneal d ialysis Both hemodialysis and peritoneal dialysis have been used success- fully in pregnancy, although randomized prospective trials to determine the optimal therapy have not been done. Early reports favored peritoneal dialysis, demonstrating greater fetal survival than with hemodialysis, although these studies were limited by small numbers of patients and the use of historical controls in some: 67% vs 20% [4] , 83% vs 42% [9] , and 63% vs 20% [10] . This benefi t has not been borne out in more recent analyses and likely refl ects improvement in outcome for pregnant patients on dialysis as a whole. The National Registry for Pregnancy in Dialysis Patients (NPDR) documented virtually identical fetal survival rates among 184 pregnancies for hemodialysis (39.5%) vs peritoneal dialysis (37%) [1] . Similar data are described by Chan and colleagues (82% vs 72%) in their review of all pregnan- Figure 13.2 Peritoneal dialysis. Table 13.3 Mode of dialysis: advantages in pregnancy. Hemodialysis Peritoneal dialysis Less work intensive for patient Stable biochemical environment No risk of peritoneal catheter - related complications Continuous fl uid removal avoids hypotension Adequate clearances late in gestation readily obtained Allows liberal fl uid intake Permits continuous insulin administration in diabetes mellitus No interruption in therapy needed after cesarean section No anticoagulation necessary Permits administration of intraperitoneal MgSO 4 in pre - eclampsia Hypertension easier to control Less severe anemia Dialysis 191 An additional benefi t of intensive dialysis is that a low level of azotemia should minimize the risk of polyhydramnios, although it is not known if this will lead to improved outcome or a decreased incidence of preterm labor. Polyhydramnios, seen in a high percentage of pregnancies, has been ascribed to the urea diuresis that normally occurs in utero due to high fetal levels of urea nitrogen, as well as to fl uid shifts that accompany intermit- tent hemodialysis [20,21] . An increased frequency of hemodialy- sis in particular limits the large interdialytic weight gains often seen in hemodialysis patients, thus avoiding hypotension and enabling better blood pressure control by minimizing that com- ponent of hypertension that is volume mediated. Modifi cation of the d ialysis p rescription With respect to hemodialysis, certain parameters of the dialysis prescription may warrant adjustment. Specifi cally, a lower sodium dialysate of 134 mEq/L is recommended due to the mild physiologic hyponatremia of pregnancy. Similarly, a bicarbonate concentration as low as 25 mEq/L may be necessary to avoid alkalemia, due to the repeated exposure to a bicarbonate dialysate and the concomitant respiratory alkalosis seen in pregnancy. Acetate dialysis is not generally recommended because it has been associated with an increased frequency of hypotension, although there are no data in pregnancy. A standard calcium dialysate can be used with both hemodialysis and peritoneal dialysis, thus ensuring a net positive calcium balance suffi cient to meet fetal requirements. Due to placental production of calcitriol, however, there is augmented gastrointestinal absorption of calcium from calcium - containing antacids; thus, serum calcium levels must be monitored to avoid hypercalcemia [22] . With both methods of dialysis, one must also monitor closely for hypokalemia, which may develop with frequent dialysis. Changes in the effi cacy of peritoneal dialysis have not been noted during pregnancy. In one patient studied there was no apparent change in peritoneal physiology or peritoneal blood fl ow as assessed by the standard peritoneal equilibration test of glucose and creatinine [23] . Similarly, Redrow and colleagues reported excellent ultrafi ltration in all patients throughout preg- nancy, and less than a one - third decrease in peritoneal solute clearance in three patients studied [10] . Dialysis and u teroplacental p erfusion Doppler fl ow velocity measurements have been performed during and after hemodialysis in an attempt to assess the effect of hemo- dialysis on uteroplacental blood fl ow. Results have been confl ict- ing, with studies reporting unchanged, worsened, and improved perfusion during dialysis as assessed by the systolic – diastolic ratio or resistance index [24 – 26] . In those patients studied, however, there was no evidence of uterine irritability or fetal distress as measured by external fetal monitoring during hemodialysis. Maternal c omplications In the past, women with severe renal disease were often advised to terminate pregnancies due to the belief that pregnancy carried sis and peritoneal dialysis. Although there are no fi rm guidelines, it is the belief of most nephrologists that a more intensive dialysis regimen is required during pregnancy to minimize fetal exposure to uremic toxins and improve outcome. This is based in part on the fact that pregnancy outcome appears to be better in those women who require initiation of dialysis due to a deterioration of renal function during pregnancy, as well as among women with signifi cant residual renal function who require dialysis before conception [9] . Infant survival as reported by the NPDR was 73% in the former group of women, although only 40% in those women who were already on dialysis at the time of preg- nancy [1] . Similar pregnancy success rates in dialysis patients were reported by Bagon and colleagues based on a review of all pregnancies in Belgium extending beyond the fi rst trimester [17] . Furthermore, pregnancy appears to be most common during the fi rst year of dialysis, presumably related to the greater residual renal function often present at the initiation of renal replacement therapy. There are reports of successful pregnancies in severely uremic patients and patients on dialysis for more than 10 years, as well as pregnancy failures in women treated with intensive dialysis. Intensive dialysis corresponds to initiation of dialysis at levels of BUN and creatinine approximately 60 – 70 mg/dL and 6 – 7 mg/ dL, respectively, with a goal of maintaining predialysis BUN levels less than 50 mg/dL and 5 mg/dL, respectively [6,9] . To maintain such low levels of azotemia in pregnancy, dialysis patients may require a signifi cant increase in total treatment time. This is espe- cially true in the third trimester when fetal urea production increases and may account for as much as 540 mg/day [18] , a 10% increase. For women on hemodialysis, daily treatments of 5 or more hours may be necessary to obtain adequate clearances late in gestation. As with hemodialysis, a patient ’ s treatment require- ments may increase markedly with peritoneal dialysis as well, especially because women in the latter half of gestation may be unable to tolerate the standard dwell volumes due to abdominal fullness. A switch to CCPD with an increased frequency of small volume exchanges and supplemental manual exchanges is often required late in gestation to obtain adequate clearance. A combi- nation of hemodialysis and peritoneal dialysis may even be indicated. Although the ideal dialysis prescription has yet to be estab- lished, the National Registry data suggest a trend towards greater infant survival and more advanced gestational age in those women receiving more than 20 hours of hemodialysis weekly [1,19] . Others have confi rmed this fi nding, although no benefi t was found in those women prescribed a higher dose of peritoneal dialysis [11] . Even though the number of weekly hemodialysis treatments had no effect on outcomes in some studies, perform- ing dialysis 4 to 6 times per week may allow for better fl uid and blood pressure management and may also decrease the risk of polyhydramnios which can lead to preterm labor and delivery. While no guidelines exist with regard to evaluating the adequacy of dialysis, a minimum combined renal and dialytic clearance of 15 mL/min is recommended. Chapter 13 192 period [31] . Three maternal deaths have been reported to date, one of which was the result of lupus cerebritis [1] . Polyhydramnios is a common fi nding, reported in between 29% and 67% of pregnancies in CRF patients [6] . This may be caused by the rapid removal of solutes during hemodialysis and shifting of free water into the amniotic space or an increased fetal osmotic diuresis because of the increased maternal urea concentration. Preterm delivery is an expected outcome in these patients as well, with the mean age of delivery being only 32 weeks gestation [32] . Early delivery may be because of spontaneous preterm labor, and polyhydramnios may contribute to this, the result of fetal distress, intrauterine growth restriction, pre - eclampsia, or placental abruption. Fetal c omplications The likelihood of fetal survival beyond the neonatal period is better than previously believed (Table 13.5 ). Surveys conducted by the EDTA [3] , the American Nephrology Nursing Association [9] , as well as a group in Saudi Arabia [5] reported a fetal viability of 20 – 30% in those pregnancies that were not electively termi- nated. The EDTA survey revealed that greater than 50% of preg- nancies resulted in spontaneous abortion [3] . Hou et al noted a comparable incidence of 54% fetal loss, including spontaneous abortion, stillbirth, and neonatal death [21] ). Virtually all infants delivered were premature, and approximately 20% were growth - restricted. When stratifi ed according to year, however, survival was greater than 50% in those pregnancies occurring since 1990. Another study supports improved neonatal survival and sum- marized the outcomes of 111 pregnancies in patients receiving chronic hemodialysis and reported that 71% (79/111) of these infants who were born survived [6] . As noted previously, polyhydramnios, possibly attributed to the fetal urea diuresis, is seen with greater frequency in renal failure and may contribute to the high incidence of prematurity. Additionally, a urea - induced diuresis following delivery may result in volume depletion in the neonatal period. Early reports failed to identify an increased incidence in con- genital anomalies [3,9] . However, the NPDR reported on 11 infants with congenital anomalies among 55 live births [1] . Not surprisingly, there was also a high proportion of infants with developmental delays or long - term medical problems docu- mented at follow - up, the latter possibly attributable to problems a high risk of maternal complications and a low success rate. Because there are defi nite risks to both the mother and the fetus as a result of CRF requiring dialysis in pregnancy, these patients should be counseled before conception if at all possible. Potential complications include an accelerated decline in renal function, accelerated hypertension, an increased risk of superimposed pre - eclampsia, polyhydramnios, worsened anemia often requiring transfusion, hemodialysis access thrombosis, and an increased incidence of abruptio placentae (Table 13.4 ). The latter cannot be ascribed solely to the use of heparin during hemodialysis because it has been seen with greater than normal frequency in patients on peritoneal dialysis as well. Pregnancy has been associated with a permanent decline in renal function in a relatively small percentage of patients with mild renal failure, defi ned by a serum creatinine of < 1.4 mg/dL. This risk may be increased signifi cantly in those women with moderate or severe renal failure, especially in the setting of uncontrolled hypertension. It is always important to rule out readily reversible causes of declining renal function, such as volume depletion, pyelonephritis, and obstruction. One report of 37 pregnant women with moderate or severe renal failure, defi ned as a serum creatinine greater than 1.4 mg/dL, demonstrated a deterioration in renal function, defi ned as greater than a 50% rise in creatinine, in 16% [27] . Five of these six women also suffered from poorly controlled chronic hypertension, and a clinical diag- nosis of superimposed pre - eclampsia was established in nearly 60% overall. Similarly, a more recent review encompassing more than 80 pregnant women with renal failure demonstrated acceler- ated hypertension in nearly 50% and an accelerated decline in renal function in more than one - third [28] . Hou reviewed these studies along with fi ve others, all of which confi rmed the increased incidence of accelerated renal failure in women with a serum creatinine greater than 1.4 mg/dL at the time of conception [29] . Severe hypertension and proteinuria were predictive of an accel- erated course in more than 20% of patients with moderate to severe renal failure due to a wide range of primary glomerular diseases [30] . Of interest, a review of pregnancy in patients with diabetic nephropathy, defi ned as nephrotic - range proteinuria and severe hypertension, failed to describe an accelerated loss of renal function during pregnancy, although nearly one - third of women had reached ESRD or died during the 3 - year follow - up Table 13.4 Renal failure and pregnancy: maternal complications. Accelerated decline in renal function Accelerated hypertension Superimposed pre - eclampsia Preterm labor Worsened anemia Hemodialysis access thrombosis Abruptio placentae Spontaneous abortion and second - trimester fetal loss Table 13.5 Renal failure and pregnancy: fetal complications. Spontaneous abortion and fetal loss (50%) Fetal/neonatal death (21 – 33%) Preterm delivery ( > 80%) Intrauterine growth restriction (20%) Polyhydramnios (29 – 67%) Maternal hypertension (35 – 72%) Dialysis 193 allow for normal fetal development. The recommended protein intake is 1.5 g/kg/day in hemodialysis and 1.8 g/kg/day in perito- neal dialysis and daily caloric intake increased to 30 – 35 kcal/kg/ day as well [29] . Increasing delivery of dialysis is recommended for worsening azotemia rather than strict protein restriction. Supplementation of water - soluble vitamins, which are removed during dialysis, is recommended, as well as supplementation with folate, zinc, and iron. Specifi cally, it is important to monitor hemoglobin and iron stores on a regular basis as oral iron supple- mentation is often insuffi cient given the increased requirements during pregnancy. Intravenous iron has been given to pregnant dialysis patients without adverse outcomes as mentioned previ- ously [35] . Standard prenatal vitamins, which may contain excess vitamin A, are best avoided. Antepartum m anagement Care during pregnancy for patients on dialysis should include a multidisciplinary team with at least initially a nephrologist and maternal - fetal medicine specialist. Given the extremely high like- lihood of a preterm delivery, it is also important to make the neonatologist and the neonatal intensive care unit aware of the patient as these infants may also demonstrate some degree of azotemia after delivery. In addition, the patient should be coun- seled on the likely complications and possibility of premature delivery. Early in pregnancy, care should be focused on dating the preg- nancy as accurately as possible, which is diffi cult given the advanced gestational age at which patients often present. Dialysis time should be increased and anemia monitored closely which will require increasing the dose of erythropoietin as discussed before. Dietary alterations, including increased folate supplemen- tation and protein intake, should also be instituted. Hypertension is a common comorbidity and is seen in up to 80% of pregnant dialysis patients [6] . Many antihypertensive medications are considered safe in pregnancy and may be uti- lized; however, the angiotension - converting enzyme inhibitors and angiotensin receptor blockers are contraindicated in preg- nancy secondary to associations with renal dysplasia, neonatal anuria, and stillbirth [36,37] . Some of the older medications that have been described to treat hypertension in pregnancy include methyldopa and hydralazine and these are acceptable to use. Even though there have been reports of concerns about intrauterine growth restriction and neonatal bradycardia and hypoglycemia with the use of some beta - blockers, the best evidence for this was associated with atenolol. Other beta blockers, such as labetalol, are recommended by some as fi rst - line therapy for treatment of hypertension in pregnancy [38] . Regardless of which medication is used, treatment of hypertension is essential during pregnancy for these patients. In addition, the physician should monitor for urinary tract infections and aggressively treat even asymptomatic bacteruria as the risk of pyelonephritis is quite high when this is present and often encountered with premature birth. Unfortunately, there is little additional long - term follow - up on infants exposed to azotemia in utero with regard to physical and intellectual development. Anemia Anemia develops during pregnancy largely due to an increase in plasma volume of 3 – 4 L without a corresponding increase in red cell mass [6] . In patients with renal failure, the picture is compli- cated by a relative defi ciency in erythropoietin production by the diseased kidneys, as well as shortened red cell survival, bone marrow suppression by uremic toxins, and possible superim- posed nutritional defi ciencies. The severe anemia that was typical of ESRD in the past is now treated successfully in most cases with recombinant human erythropoietin (rHuEpo). Furthermore, correction of the anemia of ESRD may result in return of regular menses due to resolution of hyperprolactinemia, and conception may follow [8] . Recombinant human erythropoietin has been studied in preg- nant animals at doses used clinically without apparent complica- tions. Hou reported on 11 patients with CRF treated with rHuEpo in whom no congenital anomalies were seen and no rHuEpo could be detected in the cord blood [9] . All of the women required an increase in their dose of rHuEpo, compared with prepreg- nancy, and three still required blood transfusions during pregnancy. Only one woman experienced severe hypertension complicating therapy, although several required additional anti- hypertensive medications. Additional reports have yielded similar results [1,17,33,34] . It is accepted by most obstetricians that a hemoglobin less than 6 g/dL is associated with increased perinatal mortality and maternal morbidity secondary to high - output failure. Given this fact, as well as the increased risk of bleeding complications in uremia due to platelet dysfunction, and the overwhelming likelihood of preterm delivery, the recommenda- tion for women with renal disease is an empirical 50% increase in rHuEpo dose once the pregnancy is detected, with a goal of maintaining the hemoglobin at more than 10 g/dL [9] . Most patients require oral iron supplementation or intermittent intra- venous iron, because iron defi ciency eventually develops in most patients successfully treated with rHuEpo. Although intravenous iron has been used without incident in at least 20 patients, it is generally recommended only if iron defi ciency persists despite oral therapy. Dietary g uidelines Dietary restrictions in renal failure generally consist of modest protein restriction, as well as restriction of potassium, phosphate, and sodium intake. Fluids are restricted to 1 L daily, with more liberal intake permitted in those with substantial residual urine output. In pregnancy, however, protein intake is liberalized to Chapter 13 194 general, the route of delivery should be determined by the fetal status and cesarean section reserved for normal obstetric indications. In the postpartum period, the patient must be monitored closely as the provider should anticipate signifi cant fl uid shifts in the fi rst week after delivery. Pregnancy and a cute r enal f ailure Most of the literature pertaining to dialysis in pregnancy concerns those women with CRF or ESRD. There are, however, a number of case reports of dialysis for ARF in pregnancy. Hemodialysis has been the primary form of dialysis utilized, both for ARF and for acute ingestion of toxic substances [44 – 46] . Because the inci- dence of ARF itself has fallen to less than 1% of pregnancies in developed countries, the need for acute dialysis is rare [47] . This topic is addressed in more detail in Chapter 28. Summary Although pregnancy remains uncommon in women with severe CRF or ESRD, it is nevertheless a possibility, especially with modern treatment. With intensive management by the obstetri- cian and nephrologist, the likelihood of a favorable outcome can be maximized. This will generally entail early initiation of dialysis in women with CRF or intensifi ed dialytic therapy in those already requiring renal replacement therapy. References 1 Okundaye I , Abrinko P , Hou S . Registry of pregnancy in dialysis patients . Am J Kidney Dis 1998 ; 31 : 766 – 773 . 2 Toma H , Tanabe K , Tokumoto T et al. A nationwide survey on preg- nancies in women on renal replacement therapy in Japan . Nephrol Dial Transpl 1998 ; 31 : A163 . 3 Registration Committee of the European Dialysis and Transplant Association. Successful pregnancies in women treated by dialysis and kidney transplantation . Br J Obstet Gynaecol 1980 ; 87 : 839 – 845 . 4 Gadallah MF , Ahmad B , Karubian F , Campese VM . Pregnancy in patients on chronic ambulatory peritoneal dialysis . Am J Kidney Dis 1992 ; 20 : 407 – 410 . 5 Souqiyyeh MZ , Huraib SO , Saleh AG , Aswad S . Pregnancy in chronic hemodialysis patients in the Kingdom of Saudi Arabia . Am J Kidney Dis 1992 ; 19 : 235 – 238 . 6 Reddy SS , Holley JL . Management of the pregnant chronic dialysis patient . Adv Chronic Kid Dis 2007 ; 14 ( 2 ): 146 – 155 . 7 Schwarz A , Post KG , Keller F , Molzahn M . Value of human chorionic gonadotropin measurements in blood as a pregnancy test in women on maintenance hemodialysis . Nephron 1985 ; 39 : 341 – 343 . 8 Hou SH , Orlowski J , Pahl M et al. Pregnancy in women with end - stage renal disease: treatment of anemia and premature labor . Am J Kidney Dis 1993 ; 21 : 16 – 22 . untreated. Generally, urine cultures can be done every 4 – 6 weeks and patients treated as appropriate. A new approach to the prevention of recurrent preterm deliv- ery may also have some benefi t in dialysis patients. Two random- ized trials have now demonstrated a signifi cant reduction in the incidence of recurrent preterm birth with the administration of progesterone supplementation during pregnancy [39,40] . Given the extremely high risk of preterm delivery, as well as additional evidence that dialysis may signifi cantly affect serum progesterone levels, it is not unreasonable to consider the administration of progesterone supplementation, either by injection or vaginal sup- positories, beginning around 16 – 20 weeks [19] . After a sonogram for a complete anatomic survey between 18 and 20 weeks, the patient should be followed with serial growth scans approximately every 4 weeks. Unless indicated earlier, ante- partum fetal testing should begin at 28 – 30 weeks and include twice weekly non - stress tests and at least a weekly measurement of the amniotic fl uid index (AFI). If intrauterine growth restric- tion is diagnosed, then monitoring with serial umbilical artery Doppler measurements is appropriate as well. If preterm labor occurs, then medical interventions may be undertaken in an effort to prolong the pregnancy and improve fetal outcomes. Steroids should be administered, either beta- methasone or dexamethasone, according to standard protocols. Tocolytics may be given and, before 32 weeks, indomethacin can and has been used in women with renal disease [41] . It is impor- tant to note, however, that indomethacin is generally not contin- ued for greater than 72 hours or given after 32 weeks secondary to the potential for premature closure of the fetal ductus arterio- sus and fetal anuria. Magnesium sulfate may also be used for tocolysis with a target level of 5 – 7 mg/dL considered to be thera- peutic. Care must be taken with the use of magnesium in renal failure patients because it is cleared though the kidney and the provider must monitor serum levels closely and watch for evi- dence of magnesium toxicity. Calcium channel blockers, such as nifedipine, may also be used in dialysis patients for tocolysis. Delivery is often required for pregnant patients on dialysis for intrauterine growth restriction and preterm premature rupture of membranes (PPROM). In general, if PPROM occurs at 34 weeks or later, delivery is indicated as there is no signifi cant benefi t to prolonging the pregnancy and the risk of infection is more concerning. There are no data to support preference for any difference in the mode of delivery in dialysis patients. Reported rates of cesar- ean section range from 24% to more than 60% [3,11,16,34,42] . Cesarean delivery should be performed for standard obstetric indications. In peritoneal dialysis patients requiring cesarean section, both standard and extraperitoneal approaches have been utilized [10,43] . In either case, it may be necessary to interrupt peritoneal dialysis for several days to allow healing of the abdomi- nal wall and prevent dialysate leak or hernia formation. Peritoneal dialysis can be reinitiated using smaller dwell volumes initially, with a progressive increase in volume as tolerated. If necessary, temporary hemodialysis can be performed in the interim. In Dialysis 195 29 Hou S . Pregnancy in chronic renal insuffi ciency and end - stage renal disease . Am J Kidney Dis 1999 ; 33 : 235 – 252 . 30 Jungers P , Chauveau D , Choukroun G et al. Pregnancy in women with impaired renal function . Clin Nephrol 1997 ; 47 : 281 – 288 . 31 Reece EA , Leguizamon G , Homko C . Pregnancy performance and outcomes associated with diabetic nephropathy . Am J Perinatol 1998 ; 15 : 413 – 421 . 32 Holly J , Reddy S . Pregnancy in dialysis patients: A review of outcomes, complications, and management . Semin Dial 2003 ; 16 : 389 – 402 . 33 Barth W , Lacroix L , Goldberg M , Greene M . Recombinant human erythropoietin (rHEpo) for severe anemia in pregnancies complicated by renal disease . Am J Obstet Gynecol 1994 ; 170 : 329A . 34 Scott LL , Ramin SM , Richey M et al. Erythropoietin use in pregnancy: two cases and a review of the literature . Am J Perinatol 1995 ; 12 : 22 – 24 . 35 Hou S , Firanek C . Management of the pregnant dialysis patient . Adv Ren Replace Ther 1998 ; 5 : 24 – 30 . 36 Pryde PG , Sedman AB , Nugent CE et al. Angiotensin - converting enzyme inhibitor fetopathy . J Am Soc Nephrol 1993 ; 3 : 1575 – 1582 . 37 Bhatt - Mehta V , Deluga KS . Fetal exposure to lisinopril: Neonatal manifestations and management . Pharmacotherapy 1993 ; 13 : 515 – 518 . 38 Sibai BM . Hypertension . In: Gabbe SG , Niebyl JR , Simpson J , eds. Obstetrics: Normal and Problem Pregnancies , 5th edn. Philadelphia, PA : Elsevier , 2007 : 903 . 39 DaFonseca EB , Carvalho MHB , Zugaib M . Prophylactic administra- tion of progesterone by vaginal suppository to reduce the incidence of spontaneous preterm birth in women at risk: a randomized, placebo - controlled double - blind study . Am J Obstet Gynecol 2003 ; 188 : 419 – 424 . 40 Meis PJ , Kleanoff M , Thorn E , Dombrowski MP , Sibai B , Moawad AH , Spong CY et al. Prevention of recurrent preterm delivery by 17 alpha - hydroxyprogesterone caproate . N Engl J Med 2003 ; 348 : 2379 – 2385 . 41 Reister F , Reister B , Heyl W et al. Dialysis and pregnancy – A case report and review of the literature . Ren Fail 1999 ; 21 : 533 – 539 . 42 Yasin SY , Bey Doun SN . Hemodialysis in pregnancy . Obstet Gynecol Surv 1988 ; 43 : 655 – 668 . 43 Hou SH . Pregnancy in continuous ambulatory peritoneal dialysis (CAPD) patients . Perit Dial Int 1990 ; 10 : 201 – 204 . 44 Trebbin WM . Hemodialysis and pregnancy . JAMA 1979 ; 241 : 1811 – 1812 . 45 Kleinman GE , Rodriquez H , Good MC , Caudle MR . Hypercalcemic crisis in pregnancy associated with excessive ingestion of calcium carbonate antacid (milk – alkali syndrome): successful treatment with hemodialysis . Obstet Gynecol 1991 ; 78 : 496 – 499 . 46 Devlin K . Pregnancy complicated by acute renal failure requiring hemodialysis . Anna J 1994 ; 27 : 444 – 445 . 47 Krane NK . Acute renal failure in pregnancy . Arch Intern Med 1988 ; 148 : 2347 – 2357 . 9 Hou SH . Frequency and outcome of pregnancy in women on dialysis . Am J Kidney Dis 1994 ; 23 : 60 – 63 . 10 Redrow M , Lazaro C , Elliot J et al. Dialysis in the management of pregnant patients with renal insuffi ciency . Medicine 1988 ; 67 : 199 – 208 . 11 Chan WS , Okun N , Kjellstrand CM . Pregnancy in chronic dialysis: a review and analysis of the literature . Int J Artif Organs 1998 ; 21 : 259 – 268 . 12 Jakobi P , Ohel G , Szylman P , Levit A , Lewin M , Paldi E . Continuous ambulatory peritoneal dialysis as the primary approach in the man- agement of severe renal insuffi ciency in pregnancy . Obstet Gynecol 1992 ; 79 : 808 – 810 . 13 Changs H , Miller MA , Bruns FJ . Tidal peritoneal dialysis during preg- nancy improves clearance and abdominal symptoms . Perit Dial Int 2002 ; 22 : 272 – 274 . 14 Castillo AA , Lew SQ , Smith AM , Bosch JP . Women issues in female patients receiving peritoneal dialysis . Adv Ren Replace Ther 1999 ; 6 : 327 – 334 . 15 Elliot JP , O ’ Keeffe DF , Schon DA , Cherem LB . Dialysis in pregnancy: a critical review . Obstet Gynecol Surv 1991 ; 46 : 319 – 324 . 16 Hou SH . Pregnancy and birth control in CAPD patients . Adv Perit Dial 1993 ; 9 : 173 – 176 . 17 Bagon JA , Vernaeve H , de Muylder X et al. Pregnancy and dialysis . Am J Kidney Dis 1998 ; 31 : 756 – 765 . 18 Hou SH , Grossman SD . Pregnancy in chronic dialysis patients . Semin Dial 1990 ; 3 : 224 – 229 . 19 Hou S . Pregnancy in dialysis patients: Where do we go from here? Semin Dial 2003 ; 16 ( 5 ): 376 – 378 . 20 Nageotte MP , Grundy HO . Pregnancy outcome in women requiring chronic hemodialysis . Obstet Gynecol 1988 ; 72 : 456 – 459 . 21 Hou SH . Pregnancy in women on haemodialysis and peritoneal dialy- sis . Bailliere ’ s Clin Obstet Gynaecol 1994 ; 8 : 481 – 500 . 22 Grossman S , Hou S . Obstetrics and gynecology . In: Daugirdas JT , Ing TS , eds. Handbook of dialysis . New York : Little, Brown , 1994 : 649 – 661 . 23 Lew SQ , Watson JA . Urea and creatinine generation and removal in a pregnant patient receiving peritoneal dialysis . Adv Perit Dial 1992 ; 8 : 131 – 135 . 24 Weiner Z , Thaler I , Ronen N , Brandes JM . Changes in fl ow velocity waveforms in umbilical and uterine artery following haemodialysis . Br J Obstet Gynaecol 1991 ; 98 : 1172 – 1173 . 25 Jakobi P , Weiner Z , Geri R , Zaidise I . Umbilical and arcuate uterine artery fl ow velocity measurements during acute hemodialysis . Gynecol Obstet Invest 1993 ; 37 : 247 – 248 . 26 Krakow D , Castro LC , Schwieger J . Effect of hemodialysis on uterine and umbilical artery Doppler fl ow velocity waveforms . Am J Obstet Gynecol 1993 ; 170 : 1386 – 1388 . 27 Cunningham FG , Cox SM , Harstad TW et al. Chronic renal disease and pregnancy outcome . Am J Obstet Gynecol 1990 ; 163 : 453 – 459 . 28 Imbasciati E , Ponticelli C . Pregnancy and renal disease: predictors for fetal and maternal outcome . Am J Nephrol 1991 ; 11 : 353 – 362 . 196 Critical Care Obstetrics, 5th edition. Edited by M. Belfort, G. Saade, M. Foley, J. Phelan and G. Dildy. © 2010 Blackwell Publishing Ltd. 14 Cardiopulmonary Bypass Katherine W. Arendt Mayo Clinic, Rochester, MN, USA Introduction Cardiopulmonary bypass (CPB) is a commonly used and often necessary technique during cardiac surgery. It results in signifi - cant alterations in patient physiology with virtually every organ system affected. Some of the prominent adverse effects include: (i) profound alterations in coagulation (dilution of all clotting factors, intense heparinization, platelet dysfunction); (ii) distur- bances in cardiovascular function (hypotension, non - pulsatile blood fl ow, myocardial ischemia and cardiac stunning, arrhyth- mias); and (iii) a signifi cant generalized systemic infl ammatory response. Systemic embolization of particulate material occurs, including marrow and fat spilled into the chest when the sternum is split. Air embolization also frequently occurs. Embolic phe- nomena are thought to be major contributors to the signifi cant risk of cerebrovascular accident (2 – 6%) and neurocognitive dysfunction (20 – 60%) [1 – 3] . In addition, the management of patients requiring CPB frequently includes the use of hypother- mic techniques, invasive monitoring, and the administration of a variety of cardiovascular drugs. Not infrequently, complications involving one or more major organ systems are experienced. The application of “ off - pump ” coronary artery bypass for patients with coronary artery disease has become popular to a great degree because of the inherent risks of CPB [4] . However, off - pump approaches are not available for patients with cardiac valvular surgical disease, the cardiac pathology that most often affects the pregnant patient. Closed mitral valvotomy in order to avoid cardiopulmonary bypass (CPB) in pregnancy has been widely described [5 – 7] . Likewise, more and more cardiac proce- dures are being done without requiring CPB with percutaneous modalities. However, fetal radiation exposure during fl uoroscopy remains a prohibitive factor for percutaneous techniques during pregnancy. Echocardiography imaging alone without fl uoro- scopy has been described in percutaneous balloon valvuloplasty [8,9] , but experience in this technique is limited. For now, many cardiac procedures will continue to require parturients to undergo CPB for open heart or aorta surgeries. The fi rst reports of cardiac surgery during pregnancy were published in 1952 and involved 11 closed mitral commissuroto- mies performed during pregnancy [10 – 13] . In 1958, CPB was fi rst performed on a pregnant woman [14] . Since this time, multiple cases and series have been published describing CPB in pregnancy [15] . Maternal mortality for cardiac surgeries does not appear to be affected by pregnancy, but the risk of fetal loss is signifi cant. Maternal and f etal r isks of c ardiopulmonary b ypass For understandable reasons, no well - controlled studies have been reported assessing the impact of CPB on the pregnant patient, the fetoplacental unit, or fetal outcome. In addition, many of the existing case series date back to the late 1950s and early 1960s. Approaches and techniques that cardiac patients receive continue to change. Thus, many conclusions regarding management and outcome must be viewed with caution. From the data that are available, pregnant women do well during CPB with a variable mortality rate similar to that for non - pregnant patients: 1.4 – 13.3%. Fetal loss, however, is signifi cant, with rates ranging from 16% to 38.5%. These quoted mortality rates are from multiple published series [16 – 22] . The fi rst series of CPB in pregnancy was published in 1969 and described 20 cases with a single (5%) maternal death and 7 (33%) fetal deaths [16] . The highest fetal mortality was reported in a series of 15 parturients in Mexico undergoing open heart surgery from 1972 to 1998 with 2 (13.3%) maternal deaths and 5 (38.5%) fetal deaths [22] . In the most recent literature, Weiss et al. (1998) [23] describes 59 cases of CPB in pregnancy reported in the literature from 1984 to 1996 with 3 (5%) maternal deaths and a fetal/ neonatal mortality rate of 29%. They also noted a 25% rate of premature births. Cardiopulmonary Bypass 197 Whether by thoracotomy or a percutaneous balloon technique, mitral commissurotomy instead of open valve repair or replace- ment is associated with a signifi cant likelihood of patients requir- ing additional surgery at a later date. Mangione and coworkers [28] published favorable results with only 9% of their 23 patients requiring repeat valvuloplasty after 8 years of follow - up. Fawzy and colleagues [29] reported that 16% of their patients under- going mitral balloon valvuloplasty developed restenosis over a follow - up period of 9 years. In the series of Vosloo and Reichart [5] , 22% of patients receiving closed commissurotomy required an additional cardiac surgery during a follow - up period lasting from 5 to 17 years. These data led some to recommend that parturients undergo the open valve repair or replacement requiring CPB during pregnancy. The performance of coronary artery bypass grafting (CABG) without CPB ( ‘ off - pump ’ CABG, beating heart CABG) avoids the risks of CPB. Silberman and coworkers [33] describe a case of coronary artery bypass grafting performed on a beating heart without the use of CPB on a patient at 22 weeks gestation status after spontaneous dissection of the left anterior descending artery. She subsequently gave birth to a healthy term baby. Although long - term (10 years) results of beating heart CABG are not yet known to be equal to that of traditional CABG techniques, it has gained wider acceptance. This is because studies to date indicate that off - pump CABG techniques provide complete revascularization, reduced myocardial injury, less coagulopathy, decreased transfusion requirements, higher hematocrit at dis- charge, and shorter hospital length of stay [34] . Timing c ardiac s urgery d uring p regnancy Cardiologists who care for women with heart disease of child- bearing age counsel these patients about the risks of pregnancy and optimizing their condition before conception. Such optimi- zation is not always possible, pregnancy is not always planned, and often the existence or extent of cardiac disease is unmasked by the cardiovascular changes of the pregnancy itself. Therefore, cardiologists and cardiovascular surgeons are left trying to decide if surgical intervention is going to be necessary before term deliv- ery and, if so, the optimal gestational age to intervene. Typically, in obstetric medicine, what is in the best interest of maternal health is in the best interest of her fetus. This may not be the case when deciding the timing of surgery in the parturient with a deteriorating cardiac status. In a systematic review from 1984 to 1996, Weiss et al. compared maternal and fetal outcomes in cardiac surgeries performed during pregnancy, those per- formed immediately after delivery of the neonate, and those in which the surgery was delayed until after the postpartum period [23] . Fetal mortality was greatest (about 30%) in those surgeries performed during pregnancy, with two (5%) fetal deaths when the mother underwent surgery immediately after delivery, and no fetal deaths when the mother delayed surgery until the post- partum period. In contrast, however, the maternal mortality With such signifi cant fetal death rates, it is likely that fetal morbidity is similarly high in the fetal survivors of CPB. At present, there are no long - term follow - up studies assessing the probable deleterious effects of CPB on those with fetal exposure to CPB. The confounding effects of fetal exposures to their mother ’ s cardiac disease, pharmacologic management, and possibly other cardiac interventions would make such assess- ments diffi cult. In spite of the signifi cant number of series that have been published describing mortality rates associated with CPB in preg- nancy, few correlations can be made with CPB techniques and reduction in maternal and fetal morbidity and mortality. The optimal gestational age at the time of surgery, fetal heart rate monitoring, high fl ow CPB, normothermic CPB, and possibly pulsatile CPB have all been proposed to improve outcome. Cardiac p rocedures a voiding c ardiopulmonary b ypass The physiologic changes of pregnancy involve increases in cardiac output. Because of this, left - sided obstructive lesions such as mitral or aortic valvular stenosis are more likely than other val- vular lesions to cause complications during pregnancy [24] . Likewise, most available reports of cardiac surgery during preg- nancy involve valve repair or replacement. With the signifi cant fetal risks CPB, parturients with mitral stenosis may be evaluated for candidacy for closed mitral com- missurotomy. In this procedure, CPB is not necessary. Instead, the cardiothoracic surgeon performs an anterior lateral thora- cotomy, places his or her fi nger inside the left atrium, guides a dilator across the mitral valve orifi ce and splits open the narrowed mitral valve. Early collective experience in over 500 patients undergoing closed mitral commissurotomy before 1965 was asso- ciated with maternal mortality of under 2% and fetal mortality under 10% [25] . Multiple more recent studies have confi rmed that this procedure is lower risk for the fetus than open proce- dures requiring CPB [5,18,20] . Presently, a more common technique to avoid CPB in the parturient with mitral stenosis is percutaneous balloon mitral commissurotomy [26 – 30] . Mishra and colleagues (2001) [31] reported improved hemodynamics and symptoms in 81 out of 85 severely symptomatic pregnant women with critical mitral stenosis who underwent this procedure. They noted that although the procedure was safe and generally effective, mitral regurgita- tion increased by 1 – 2 grades in 18 of the 85 patients. Abouzied and coworkers (2001) [32] reported similar results in 16 pregnant women with severe mitral stenosis who underwent balloon mitral commissurotomy. They also reported no immediate detrimental effects of radiation exposure related to fl uoroscopy on the fetuses. In order to avoid fetal radiation exposure, echocardiography imaging alone without fl uoroscopy has been described for this procedure [8,9] . Few centers offer this technique. Chapter 14 198 rate depending upon the lesion. Therefore, if it is determined that a parturient may not survive into the third trimester, early second trimester may be the ideal time to perform surgery. This prevents further deterioration of cardiac status but exposes the fetus to anesthesia and CPB after organogenesis has occurred. Others suggest that cardiac surgery is best done between 24 and 28 weeks gestation after the attainment of fetal viability. With this timing, neonatal intensive care facilities should be available and if fetal distress is detected, cesarean delivery could occur perioperatively. Cesarean delivery during CPB should not occur because of the signifi cant bleeding risk to the mother during heparinization. Successful cesarean delivery just before CPB has been described [16,40] . Uteroplacental p erfusion and c ardiopulmonary b ypass Uteroplacental blood fl ow (UPBF) is the major determinant of oxygen and other essential nutrient transport to the fetus. A direct correlation between uterine blood fl ow (UBF) and fetal oxygen- ation has been demonstrated in both animal models and humans [41,42] . UPBF is derived primarily from uterine arteries, with a smaller contribution (of unknown signifi cance) coming from the ovarian arteries. The uterine arteries are branches of the internal iliac arteries. Uterine artery blood fl ow (UABF) increases two - to threefold in pregnancy and can represent up to 12% of the cardiac output. Increases in UBF during pregnancy are due to both physi- cal (increased diameter of the uterine artery) and physiological (decreased responsiveness of the uterine artery to endogenous circulating vasoconstrictors) mechanisms. Selective uterine artery relaxation during pregnancy may be the result of vasodilators released from its endothelium, such as PGI 2 or nitric oxide, or local hormonal actions, which diminish the activity of certain intracellular enzymes that mediate vasoconstriction. Therefore, under normal circumstances during pregnancy, the uterine arteries are maximally dilated and there is no autoregula- tion of UABF. Systemic hypotension results in vascular dilation to maintain blood fl ow for autoregulated organs such as the brain and kidneys. In contrast, the placental vasculature cannot further vasodilate in response to hypotension, and decreased uteropla- cental perfusion and, if signifi cant, fetal hypoxia results. Fetoplacental suffi ciency is related to fetal heart rate (FHR) with acute insuffi ciency resulting in fetal bradycardia and long - term insuffi ciency with subsequent fetal acidosis resulting in fetal tachycardia with minimal beat - to - beat variability on FHR tracing. An example of FHRs throughout CPB is provided in Figure 14.1 . The onset of CPB is typically characterized by fetal bradycardia, while the conclusion of CPB demonstrates fetal tachycardia with minimal beat - to - beat variability [43,44] . The cause of this initial fetal bradycardia is thought to be sec- ondary to placental hypoperfusion because it has been found reversible in most cases by increasing the perfusion rate. Other theories for this initial fetal bradycardia have included maternal increased when the surgery was delayed until after birth. Therefore, it seems that the fetus benefi ts most from delaying maternal cardiac surgery until after birth, but the mother may benefi t from an earlier intervention, while still pregnant. Clearly, these retrospective data may be confounded by the fact that the sickest parturients were unable to wait for surgery and had to be treated earlier. Nonetheless, the challenge of determining the optimal timing for a deteriorating parturient is diffi cult and this study illustrates the consequences of this important clinical judg- ment: decreasing maternal risk by intervening early may result in fetal demise while delaying until after delivery may result in maternal death. In determining the optimal gestational timing for cardiac surgery, the effects of general anesthesia need to be considered separately from the effects of CPB. The most thorough evaluation of the risks of all types of anesthesia and surgery during preg- nancy retrospectively evaluated a population of 720 000 pregnant women who underwent 5405 surgical procedures [35] . The inci- dence of congenital malformations or stillbirths was not increased in the offspring in the women who underwent surgery, regardless of gestational age at the time of surgery. The incidence of prema- turity, low - birth - weight infants and the rate of infant death within 168 hours of birth was slightly increased. This increase, however, was not linked to the gestational age at the time of surgery. Further, patients who require surgery may have underly- ing illness that affects the health of their pregnancy. This con- founds the results making it diffi cult to determine the singular risk of surgery during pregnancy. Very few of the cases in this series involved CPB. Therefore, although we can state that anes- thesia at any time during pregnancy is probably safe, the risks of fetal exposure to CPB at various times during gestation are less clear. No relationship between gestational age at the time of CPB surgery and fetal morbidity and mortality can be conclusively determined at this time. There is, however, a case report of a parturient undergoing mitral valve surgery at the 6th week of gestation with fetal hydrocephalus detected at 18 weeks gestation by ultrasound [36] . Some quote this case along with a case from the 1960s as reason to avoid surgery requiring CPB during the fi rst trimester [37,38] . A case report of fetal hydrocephalus and hydrops has also been described after CPB at 19 weeks gestational age, illustrating that the second trimester is not free from fetal risk. [39] In retrospective series of parturients undergoing CPB, fetal mortality has been described during every trimester of gesta- tion [16 – 18,22,23] . Therefore, although the risks of anesthesia and CPB during fi rst trimester and organogenesis would theoreti- cally increase fetal risks, there are no data to support this theory. None the less, many anesthesiologists, cardiologists, obstetricians and cardiothoracic surgeons recommend that surgery, especially surgery requiring CPB, be delayed until after organogenesis during the fi rst trimester of pregnancy. During late second trimester, the cardiac output of the parturi- ent peaks. As a result, if the parturient is doing poorly at the beginning of the second trimester, she will likely further deterio- . disease: predictors for fetal and maternal outcome . Am J Nephrol 1991 ; 11 : 353 – 362 . 196 Critical Care Obstetrics, 5th edition. Edited by M. Belfort, G. Saade, M. Foley, J. Phelan and G. Dildy Standard prenatal vitamins, which may contain excess vitamin A, are best avoided. Antepartum m anagement Care during pregnancy for patients on dialysis should include a multidisciplinary team. : 456 – 459 . 21 Hou SH . Pregnancy in women on haemodialysis and peritoneal dialy- sis . Bailliere ’ s Clin Obstet Gynaecol 1994 ; 8 : 481 – 500 . 22 Grossman S , Hou S . Obstetrics and gynecology