Case Report Bullous Cellulitis Caused by Serratia marcescens Curtis L Cooper, MD;* Marni Wiseman, MD;* and Robert Brunham, MD*+ Bullous cellulitis is a distinctive form of cellulitis most often caused by beta hemolytic streptococci This report describes a case of bullous cellulitis caused by Serratia marcescens in an elderly diabetic woman with peripheral vascular disease A discussion of this ubiquitous, nosocomial pathogen follows CASE REPORT A 69;year-old First Nations female with type II diabetes mellitus, coronary artery disease, and peripheral vascular disease presented to a northern medical station following a l-week history of pain and erythema of her left foot that was associated with fever and chills Forty-eight hours after onset of the fever, two large, purple bullae developed in the area of cellulitis on the dorsum of the left foot and the anterior surface of the distal left leg At that time, empirical treatment with intravenous clindamycin and gentamicin was initiated The patient denied further fever; however, she continued to experience significant discomfort in her left foot This pain was partially relieved with dependency of the foot Concerns over poor response to therapy prompted transfer to the Health Sciences Centre, Winnipeg Past pertinent medical history included removal of an ingrown and chronically infected left great toenail month prior to presentation The patient received no perioperative antimicrobial therapy; however, she did soak her toe in chlorhexadine, an antiseptic solution Additionally, she complained of intermittent claudication of both her calves, but denied rest pain She did not have a past history of diabetic foot ulcers Upon presentation, the patient was not in distress, but complained of prominent left foot pain Examination of the left limb revealed two tense, purple bullae, each measuring approximately 10 X 15 cm (Figure 1) One was located on the dorsum of the foot and the other on the dis tal medial ankle A limited area of cellulitis surrounded the bullae The base of the bullae was erythematous This region was not purulent or foul smelling Neither crepitus nor lymphangitis was noted The great toenail was absent The foot was cool to palpation, with diminished pulses distal to the popliteal artery Mild pedal edema was present bilaterally Laboratory examination revealed a leukocyte count 17.1 X 109/L with a left shift of 7%, and an erythrocyte sedimentation rate of 99 mL/hr A femoral angiogram revealed extensive peripheral vascular disease with diminished blood flow distal to the popliteal artery and no evidence of arterial thrombosis Lower limb pressure studies of the left leg indicated severe obstruction, with low distal perfusion and a halluxbrachial systolic pressure index of 0.14 Studies of the right lower limb indicated only moderate obstruction of blood flow in the main arteries with good distal perfusion and a ratio of 53 Doppler ultrasound did not reveal a deep vein thrombosis and radiography did not demonstrate any bony abnormalities Blood cultures were negative Low viscosity, translucent, red-colored fluid was aspirated from both bullae Gram stain of the fluid revealed 10 polymorphic leukocytes per high-power field and 10 to 100 gram-negative bacilli per high-power field No red blood cells were noted The sole organism isolated after aerobic and anaerobic culture from both bullae was S.marcescens The antibiogram indicted resistance to ampicillin (MIC >16 pg/mL), cefazolin (MIC >16 pg/mL), and cefuroxime (MIC >16 ug/mL) The organism was sensitive to third generation cephalosporins, aminoglycosides, and ciprofloxacin (MIC < pg/mL) Following identification of S marcescens the patient was started on a IO-day course of ciprofloxacin Resolution of the cellulitis began at day of treatment, and the patient was discharged to the northern medical station At the northern medical station, it was noted that despite antimicrobial therapy and meticulous wound care, the left ankle became progressively ischemic and ultimately progressed to vascular gangrene The patient underwent a left below knee amputation month after her discharge from the Health Services Centre Her postoperative course was uncomplicated DISCUSSION *Department of Internal Medicine, and +Department biology, University of Manitoba, Winnipeg, Manitoba, Address correspondence Building, 730 William 36 of Medical Canada Micro- to Dr Curtis L Cooper, Room 543 Basic Sciences Avenue, Winnipeg, Manitoba, Canada, R3E OW3 Bullous cellulitis occurs most often with beta hemolytic streptococci (Streptococcus pyogenes) infection and less commonly with infection due to Staphylococcus aureus Secondary bullous cellulitis complicating gram-negative Bullous Cellulitis / Cooper et al Figure Examination on the dorsum (right) of the left foot revealed tense, purole bullae, bacteremia rarely has been documented Organisms isolated include Pseudomonas aeruginosa, Escberichia coli, Aeromonas hydrophila, Morganella morganii, Enterobacter cloacae, Vibrio vulnajiicus, and Salmonella enteritidis.‘-’ Primary cases due to the above organisms have not been documented To the authors’ knowledge, S.marcescens previously has been associated with bullous cellulitis in a single case report.’ This patient was an immunocompromised male who developed a polymicrobial infection at the site of a biopsy The present case is the first report of bullous cellulitis caused solely by S marcescens This association is supported by the fact that S marcescens alone was isolated from three separate aspirations The combination of fever, local discomfort, appearance of the region surrounding the bullae, and the Gram stain suggest that infection, rather than colonization, was present The patient’s stable clinical presentation and negative blood cultures suggest that the bullous cellulitis was primary and not secondary to bacteremic seeding of S marcescens from another source measuring approximately 10 X 15 cm on the distal medial ankle 37 (left) and Risk factors for infection with S marcescens include chronic debilitating disease, diabetes mellitus, corticosteroid use, recent therapy with broad-spectrum antibiotics, indwelling catheters, mechanical ventilation, and tracheostomy’ This patient had had diabetes for over 20 years and had been on antibiotics for cellulitis of the left great toe month previously Serratia marcescens is a ubiquitous organism found in both soil and water environments Its ecology in moist environments results in its frequent isolation as a contaminant of ventilation equipment, tracheostomy tubes, and indwelling catheters Without appropriate precautions, normally sterile hospital solutions can become contaminated with this organism Potential sources of infection include peritoneal dialysis fluid, enteral feeling solutions, and antiseptic solutions lo-l3 Serratia marcescens cellulitis has been reported in a patient on hemodialysis that was presumed to be secondary to dialysate contamination.‘* In the present case, the patient gave a history of soaking her ingrown great toe in chlorhexidine antisep- 38 International Journal of Infectious Diseases / Volume tic solution approximately weeks prior to her presentation with bullous cellulitis It is speculated that her colonization and subsequent infection may have been attributable to contamination of her antiseptic solution Unfortunately, the antiseptic solution could not be obtained for culture Interestingly, S.marcescens is known to be resistant to chlorhexidine.12 Multidrug resistance is a well-known feature of S.marcescens Serratia possesses a chromosomal B-lactamase that necessitates high concentrations of B-lactamase inhibitors to achieve bacteriocidal levels.15 Quinolone resistance has been attributed to outer membrane protein alterations that result in diminished antibiotic permeability l6 Aminoglycoside resistance also has been reported Both aminoglycosides and third generation cephalosporins are regarded as first line therapy for Serratia infections.17a1s Oral ciprofloxacin is also recommended for serious soft tissue infections caused by this organism.r9 The patient presented here was initially treated with gentamicin, resulting in resolution of her fever and chills Subsequently, the gentamicin was discontinued and therapy with oral ciprofloxacin was initiated to complete a therapeutic course of 14 days This decision was guided by the patient’s overall satisfactory condition, the antibiogram profile of the organism, and the advantages of an oral route of administration The diagnostic differential of bullous cellulitis is broad To the authors’ knowledge, this patient represents the first case report of bullous cellulitis caused solely by S marcescens It is postulated that her exposure to the organism occurred while she soaked her foot in a contaminated antiseptic solution Diabetes and recent antibiotic use were additional contributing risk factors Empirical therapy with gentamicin, followed by ciprofloxacin resulted in resolution of the bullous cellulitis, however, vascular compromise of the leg necessitated subsequent amputation despite successful initial treatment REFERENCES Forkner GE, Freei E, Edcomb JH, et al Pseudomonas septicemia: observations on twenty-three cases Am J Med 1958; 25~877-889 3, Number 1, July-September 1998 Fisher K, Berger BW, Keusch GT Subepidermal bullae secondary to Escherichia coli septicemia Arch Dermatol 1974; 100:105-106 Francis YF, Richman S, Hussain S, et al Aeromonas hydrophila infection NY State J Med 1982; 82:1461-1464 Bagel J, Grossman ME Hemorrhagic bullae associated with Morganella morganii septicemia J Am Acad Dermatol 1985; 12:575-576 Livingston W, Grossman ME, Garvey G Hemorrhagic bullae in association with Enterobacter cloacae septicemia J Am Acad Dermatol 1992; 27:637-638 Wickboldt LG, Sanders CV Vibrio vulnificus infection J Atn Acad Dermatol 1983; 9:243-251 Wolinsky S, Grossman ME, Walther RR, et al Hemorrhagic bullae associated with SaEmoneZZa septicemia Arch Dermatol 1974; Il0:105-106 Bonner MJ, Meharg JG Jr Primary cellulitis due to Serratia marcescens JAMA 1983; 250:2348-2349 Crowder JG, Gikley GA, White AC Serratia marcescens bacteremia Arch Intern Med 1971; 128:247-253 10 Connacher AA, Old DC, Phillips G, et al Recurrent peritonitis caused by Serratia marcescens in a diabetic patient receiving continuous ambulatory peritoneal dialysis J Hosp Infect 1988; 11:155-160 11 Oie S, Kamiya A, Hironaga K, Koshiro A Microbial contamination of enteric feeding solution and its prevention Am J Infect Control 1992; 20:202-205 12 Bosi C, Davin-Regli A, Charrel R, et al Serratia marcescens nosocomial outbreak due to contamination of hexetidine solution J Hosp Infect 1996; 33:217-224 13 Nakashima AK, Highsmith AK, Mat-tone WJ Survival of Serratia marcescens in benzalkonium chloride and in multipledose medication vials: relationship to epidemic septic arthritis J Clin Microbial 1987; 25:1019-1021 14 Bornstein PE Ditto AM, Noskin GA Serratia marcescens cellulitis in a patient on hemodialysis Am J Nephrol 1992; 12:374-376 15 Richmond MM, Sykes RB The beta-lactamases of gramnegative bacteria and their possible physiological role Adv Microb Physiol 1973; 9:31-38 16 Goldstein nW, Gutman L, Williamson R, et al In vivo and in vitro emergence of simultaneous resistence to both betalactam and aminoglycoside antibiotics in a strain of Serratia marcescens Ann Microbial (Paris) 1983; 134A:329-337 17 Sanford JI: Gilbert DN, Sande MA Guide to antimicrobial therapy 26th ed Vienna, VA: Antimicrobial Therapy, Inc 1996 18 Bergin CJ, Phillips P, Chan RM, et al Treatment of Pseudomonas and Serratia infections with ceftazidime J Antimicrob Chemother 1985; 15:613-621 19 Nix DE, Cumbo TJ, Kuritzky P, et al Oral ciprofloxacin in the treatment of serious soft tissue and bone infections Efficacy, safety, and pharmacokinetics Am J Med 1987; 82:146-153 ... biopsy The present case is the first report of bullous cellulitis caused solely by S marcescens This association is supported by the fact that S marcescens alone was isolated from three separate... To the authors’ knowledge, this patient represents the first case report of bullous cellulitis caused solely by S marcescens It is postulated that her exposure to the organism occurred while... Il0:105-106 Bonner MJ, Meharg JG Jr Primary cellulitis due to Serratia marcescens JAMA 1983; 250:2348-2349 Crowder JG, Gikley GA, White AC Serratia marcescens bacteremia Arch Intern Med 1971;