Amikacin Therapy for Severe Gram-Negative Sepsis Emphasis on Infections with Gentamicin-Resistant Organisms FRANCIS P. TALLY, M.D., THOMAS J. LOUIE, M.D., WILLIAM M. WEINSTEIN, M.D., JOHN G. BARTLETT, M.D., and SHERWOOD L. GORBACH, M.D., Sepulveda, California
Amikacin (BB-K8) is a semisynthetic derivative of kanamycin which is active in vitro against many gentamicin-resistant Gram-negative bacilli. Twenty-three patients with 25 serious Gram-negative infections were treated with this new aminoglycoside. Twelve infections involved organisms that were resistant to gentamicin. Twenty patients satisfied the criteria for bacteriological and clinical cure. This included 11 of the 12 infections involving gentamicin-resistant Gram-negative bacilli. In 4 urinary tract infections there was a good clinical response, but routine follow-up urine cultures at 30 days were positive. One patient failed on amikacin therapy. Eighth nerve toxicity was detected in two patients. These results indicate that amikacin is effective in the treatment of serious Gram-negative infections and is particularly useful in those involving resistant organisms. Further studies are indicated to evaluate ototoxic potential.
have reported the emerging resistance of Enterobacteriaceae and Pseudomonads to gentamicin (Garamycin®*), a widely used antibiotic for serious Gram-negative infections (1-4). This changing pattern of antimicrobial susceptibility serves as a constant stimulus for the development of new antimicrobial agents and for the modification of existing drugs. The challenge is heightened by the existence of at least nine bacterial enzymes that are known to inactivate aminoglycoside antibiotics (5). Amikacin (BB-K8t) is a new semisynthetic derivative of kanamycin sulfate (Kantrex®t) obtained by acylation with an amino hydroxy butyric acid at the C-l amino group of the 2-deoxystreptamine moiety (Figure 1) (6). This new compound is a poor substrate for three of the four enzymes that inactivate gentamicin and for two of the three that inactivate tobramycin (Tobracin®t) (7). In-vitro susceptibility studies show that amikacin possesses good activity against Enterobacteriaceae and Pseudomonads, including many gentamicin-resistant microorganisms (8). The present study was undertaken to evaluate the SEVERAL LABORATORIES
clinical efficacy and toxicity of amikacin in the treatment of serious Gram-negative infections. Materials and Methods PATIENTS
The criterion for admission to the study was that all patients had signs of Gram-negative infection that indicated the need for antibiotic treatment and could be used as parameters of efficacy. These included fever, hypotension, chills, urinary-tract symptoms, and positive cultures. Additionally, an attempt was made to study infections involving resistant organisms according to the following guidelines: [1] previous isolation of Gramnegative pathogens that were resistant by in-vitro testing to conventional antimicrobials including penicillins, cephalosporins, and gentamicin; or [2] before the availability of bacteriological results, a clinical setting that suggested resistant bacterial pathogens. This category included patients with hospital-acquired infections and those with persistent sepsis despite several days of antimicrobial treatment with conventional agents. Informed consent was obtained from all patients or next-of-kin before amikacin treatment. MICROBIOLOGICAL STUDIES
Blood cultures were collected from all patients before, during, and after therapy. Patients with urinary-tract infections had quantitative urine cultures obtained before, during, and 30 days following therapy; bacterial isolates were considered pathogens if counts were greater than lOVml. Other bacterio-
* Schering Corporation, Bloomfield, New Jersey. t Bristol Laboratories, Syracuse, New York. t Eli Lilly & Co., Indianapolis, Indiana. • From the Infectious Disease Section, Veterans Administration Hospital, Sepulveda, California, and the UCLA School of Medicine, Los Angeles, California.
484
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
Figure 1 . Structural configuration of amikacin and kanamycin. Annals of Internal Medicine 83:484-488, 1975
logical studies included aspirates of closed spaces in appropriate situations. Nonfermentative Gram-negative bacilli were identified by the methods of Pickett and Pedersen (9). Other Gramnegative bacilli were identified according to Edwards and Ewing (10). Antimicrobial susceptibility of the isolates was initially tested by standardized disc-diffusion methods (11). Minimal inhibitory concentrations (MIC) and minimal bactericidal concentrations (MBC) were subsequently tested by broth dilution with microtiter techniques (12). The test medium was MuellerHinton broth; the inoculum was a 1-to-100 dilution of a 6-h broth culture to give a final concentration of approximately 103"4 colony forming U/ml. This inoculum was delivered to microtiter wells and the plates were incubated overnight in air at 37 °C. The MIC was the lowest concentration of antibiotic showing no macroscopic growth. The MBC was the lowest concentration showing 5 or less colonies on subculturing 0.05 ml from the microtiter wells to antibiotic free Mueller-Hinton agar.
Table 1 . Bacteriology of 25 Infections Treated with Amikacin
Organisms
Total Isolates
Blood Culture Isolates
Escherichia coli Pseudomonas aeruginosa Proteus rettgeri Serratia marcescens Proteus mirabilis Klebsiella species Providencia stuartii Proteus morganii Pseudomonas species Pseudomonas maltophilia Enterococcus Total
7 7 5 4 4 4 2 1 1 1 1 37*
3 2
1 1 1
8t
* Six patients had 2 isolates, 3 patients had 3 isolates. t One patient had 2 isolates.
DRUG DOSAGE
Amikacin therapy was initiated in doses of 500 mg every 12 hours. Patients with renal impairment received modified dosage according to established nomograms for aminoglycosides based on serum creatinine values (13, 14); blood levels of amikacin were used after initiating treatment to determine subsequent doses. In all cases, amikacin was administered as a single agent and previous antimicrobial agents were discontinued. SERUM LEVELS
Serum amikacin levels were measured by the agar well microbiological method using Staphylococcus epidermidis American Typed Culture Collection #27626 as the assay organism (15). Levels were obtained on the second day, and on the fifth or sixth day of therapy. Blood for peak values was drawn at 1 h and for troughs at 11 Vi h after intramuscular injection. MONITORING TOXICITY
Complete blood count with differential leukocyte count, urinalysis, serum creatinine, and liver function studies were obtained at least every 3 days during treatment. Audiometry was done whenever feasible before and after therapy with a Beltone Model 12-D portable audiometer* or I.A.C. 1204 series sound suitet. Results were kindly interpreted by Drs. D. Svihovec and W. Hanson of the Audiology and Speech Pathology Service of the Sepulveda Veterans Administration Hospital. PATIENT EVALUATION
Patients were considered cured if they became afebrile, all signs of infection cleared, and follow-up cultures showed eradication of the original pathogen. Patients who died subsequently of noninfectious diseases were considered cured if there was no evidence of the original infection at autopsy. Infections were judged drug failures if signs of infection persisted, follow-up cultures were positive, or alternative antimicrobial agents were required during the course of treatment. Results
Twenty-three patients with 25 infections fulfilled the criteria for admission to the study. All were men, reflecting the population of a Veterans Administration hospital. Twenty infections (18 patients) originated in the urinary tract; underlying conditions were obstructive uropathy in 9, neurologic disorders requiring suprapubic or indwelling Foley catheters in 7, nephrolithiasis in 3, and a sigmoid bladder following resection of a bladder carcinoma in 1. Three patients had septic arthritis; 2 had acute myelo* Beltone Electric Corporation, Chicago, Illinois. t Industrial Acoustics Co., Bronx, New York.
monocytic leukemia, and a third had a plastic knee prosthesis. One patient with chronic renal failure had an infection in the arteriovenous shunt. The final patient had an intraabdominal abscess following laparotomy. Fever was present at the onset of the study in 23 of the 25 infections; the mean peak rectal temperature was 39.4 °C. The 2 afebrile patients included 1 with septic arthritis, and a second with a urinary tract infection and the acute onset of hypotension. Additional signs of sepsis included leukocytosis (peripheral leukocyte count greater than 10 000/mm 3 ) in 20, hypotension (systolic blood pressure less than 90 mm Hg) in 9, and chills in 8. Bacteriologic studies in the 25 infections yielded a total of 37 pathogens; six patients had 2 isolates and three had 3 isolates from infected sites (Table 1). Escherichia coli and Pseudomonas aeruginosa were the most frequent organisms encountered. Proteus rettgeri, Serratia marcescens, Pr. mirabilis and Klebsiella were also common isolates. Seven patients had bacteremia, including one individual with both E. coli and Klebsiella in blood cultures. Bacteremic episodes occurred in 3 patients while they were receiving gentamicin before treatment with amikacin. Amikacin serum levels were obtained during treatment in all patients. The geometric mean peak level one h after injection was 25.0 /xg/ml with a range of 9.5 to 66 fig/ml. The high peak level was in a patient receiving hemodialysis. The mean geometric trough level was 3.5 /xg/ml with a range of 1.1 to 21 /xg/ml. The high trough level of 21 /xg/ml was in a patient with leukemia who had pseudomonas bacteremia, hypotension, and acute renal insufficiency. Based on mean achievable blood levels, microbial resistance was defined as an MIC greater than 16 /xg/ml for amikacin and greater than 8 tig/ml for gentamicin (16). Nine organisms were resistant to gentamicin (Table 2 ) . Four additional isolates (2 Pr. mirabilis, 1 Ps. aeruginosa, and 1 E. coli) had MIC's to gentamicin of 4 /xg/ml and were considered only intermediately susceptible. All but one of these resistant and marginally susceptible organisms were susceptible to amikacin. The exception was a Ps. maltophilia resistant to both gentamicin and amikacin that was recovered in a polymicrobial urinary tract infection. Tally et al. • Amikacin
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
485
Table 2. Resistant Organisms
Patient
Isolates
Source
Minimum Inhibitory Concentration Genta- Amikacin micin
culture of the joint fluid yielded Ps. aeruginosa susceptible to gentamicin. Despite adequate levels of gentamicin, purulent drainage continued and a gentamicin resistant isolate of Ps. aeruginosa was cultured. He was subsequently treated with amikacin, 500 mg, twice daily for 24 days. The drainage ceased, and the wound closed without removal of the prosthesis. His joint remains free of infection 1 year following treatment.
tig/ml
1 2 3 8 9 10 13 14 17 18 20 21 23
Pseudomonas aeruginosaKnee Aspirate Urine Serratia marcescens Proteus mirabilis Urine Pseudomonas aeruginosaBlood Escherichia coli Blood Proteus rettgeri Urine Proteus rettgeri Urine Urine Serratia marcesens Proteus mirabilis Urine Proteus rettgeri Urine Providencia stuartii Urine Pseudomonas maltophiliaUrine Serratia marcescens Urine
32 >64 4 4 4 32 36 64 4 64 64 >64 >64
PATIENT 7
0.5 4 16 1 8 16 16 8 4 4 16 >64 2
Five of the resistant isolates were susceptible to 128 fig/ ml or less of carbenicillin (Geopen®*) including 2 Pr. mirabilis, 1 Ps. aeruginosa, 1 Pr. rettgeri, and 1 Providencia stuartii. Disc-diffusion testing showed that the same 2 Pr. mirabilis were susceptible to cephalothin (Keflin®); another Pr. mirabilis and 1 Pr. rettgeri were inhibited by kanamycin. However, the other gentamicin-resistant isolates were resistant to these as well as other antibiotics. Evaluation of amikacin treatment in the 25 serious Gram-negative infections showed that 20 patients satisfied the criteria for clinical and bacteriological cure. Twelve of these infections involved bacteria that were resistant, or intermediately susceptible, to gentamicin; 6 had been treated with gentamicin and failed to respond—thus confirming the in-vitro data. In the other 6 cases, amikacin was used for initial chemotherapy. Eleven of the 12 gentamicin-resistant infections were cured with amikacin therapy. Three additional patients had persistent bacteremia during gentamicin therapy, although the bloodculture isolate was sensitive to gentamicin by in-vitro testing. In each instance the signs of sepsis cleared and blood cultures were negative following substitution of therapy with amikacin. Four infections were characterized by a good clinical response, but follow-up cultures were positive. All were urinary-tract infections in patients with underlying structural defects. In each instance, bacteruria was eradicated during treatment and in the 7-day follow-up urine culture. However, urine cultures at 30 days were again positive for the original organism. Illustrative Cases PATIENT 1.
A 55-year-old man with severe degenerative arthritis of the left knee underwent a total knee replacement with insertion of a plastic prothesis. He received prophylactic cephalothin during and following surgery. Postoperatively, he developed erythema, swelling, and pain at the incision. Sutures were removed, and * J. B. Roerig and Co., New York, New York. t Eli Lily & Co., Indianapolis, Indiana. 486
A 74-year-old man with parkinsonism and obstructive uropathy developed a fever of 40 °C. Both blood and urine cultures yielded Prov. stuartii that was resistant to gentamicin (MIC of 64 fig/ml). He was treated with amikacin, 500 mg, twice daily for 7 days. He defervesced in 36 h, and repeat blood and urine cultures were sterile. PATIENT 8
A 75-year-old man with acute myeloblasts leukemia developed fever to 40 °C, confusion, and hypotension. Laboratory data showed a leukocyte count of 1800/mm3 (17% neutrophils, 67% lymphocytes, 15% monocytes, and 1% eosinophils) and a platelet count of 16 760/mm3; the bone marrow showed many blast forms. Blood and urine cultures grew Ps. aeruginosa resistant to gentamicin. Following amikacin treatment, 500 mg, every 12 hours, he had a prompt defervescence with negative blood and urine cultures. During therapy the serum creatinine increased from 0.9 to 2.3 mg/100 ml. After 9 days of treatment, he had acute respiratory distress secondary to pulmonary edema and died. Autopsy showed acute myelocytic leukemia and pulmonary edema; there were no signs of residual infection. PATIENT 12
A 55-year-old man with renal failure on hemodialysis for 4 years developed swelling, erythema, and a purulent discharge at the arteriovenous shunt. Blood and wound cultures yielded Ps. aeruginosa susceptible to gentamicin. Despite 5 days of gentamicin treatment, he remained febrile with positive blood cultures. Therapy was changed to amikacin (3 doses of 500 mg during 72 hours) with rapid resolution of fever and clearing of the bacteremia. The infected shunt was then removed and antimicrobial treatment was discontinued. Blood levels of amikacin after the second dose showed a peak of 66 fig/vol with a trough of 4.5 fig/vol. Audiograms before therapy showed a mild highfrequency hearing loss, and there was no change in repeat studies 1 week following therapy. PATIENT 13
A 49-year-old man with advanced multiple sclerosis and a neurogenic bladder requiring suprapubic bladder drainage developed three separate episodes of Gram-negative sepsis. The first infection was characterized by flank pain, chills, fever of 40 °C, and a leukocytosis of 20 500 leukocytes/mm3. Urine culture yielded a gentamicin resistant Pr. rettgeri (MIC, 32 jitg/ml). Treatment with amikacin produced a prompt defervescence and clearing of bacteruria. However, urine culture at 30 days was again positive for the same organism. Two and 4 months later he developed similar signs of sepsis with a gentamicin-resistant Pr. rettgeri and Pr. mirabilis (MIC of 4 /ig/ml to gentamicin), respectively. On each occasion he had a prompt clinical response to amikacin. PATIENT 18
One patient failed on amikacin therapy. A 62-year-old man with atrial fibrillation and diabetes was admitted comatose with a left hemiparesis, retinal exudates, and papilledema. A Pr. mirabilis urinary tract infection was treated with ampicillin, 500 mg, intravenously every 6 h. Eight days later he developed a fever of 39.6 °C and leukocytosis of 35 100 leukocytes/mm3. At this time urine culture yielded a gentamicinresistant P. rettgeri and blood cultures were positive for E. coli; both strains were sensitive to amikacin. Amikacin therapy was initiated, but the patient continued to be febrile, and he died
October 1975 • Annals of Internal Medicine • Volume 83 • Number 4
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
40 hours later. Post-mortum examination showed a massive intracranial hemorrhage. The primary cause of death was attributed to the cerebral hemorrhage, but Gram-negative septicemia was considered a contributing factor. Pretreatment and posttreatment audiograms were available for 20 patients. Eighteen showed no change. However, one patient had a 15-dB threshold shift at 8 kHz after three courses of amikacin therapy totalling 24 g during a 6-month period. Another patient had a 20-dB shift in the midfrequency range. This individual received 7.25 g of amikacin during 10 days; the initial serum creatinine was 1.8 mg/100 ml, which decreased to 1.1 m g / 100 ml; but creatinine clearance remained unchanged (26 m l / m i n ) . Peak serum levels were only 27 and 21 ^g/ml, but trough levels were somewhat elevated to 12 and 7 fig/ml Renal function as measured by serum creatinine levels remained unchanged in 20 of 23 patients. One patient had an elevated serum creatinine associated with septic shock. He died of his underlying leukemia after 9 days of amikacin therapy. The cause of the renal insufficiency was thought to be secondary to an episode of hypotension, but the role of amikacin in this complication is unknown. Two additional patients with obstructive uropathy had a transient rise and fall in their serum creatinine levels during amikacin therapy. N o other complications were noted. Discussion
Amikacin (BB-K8) is a derivative of kanamycin with pharmacokinetics similar to those of the parent compound (17). It is active in vitro against many Enterobacteriaceae and Pseudomonads ( 7 ) . Comparative studies of various aminoglycosides have shown that amikacin is less active on a weight basis than gentamicin and tobramycin, but this disadvantage is counterbalanced by the higher achievable serum levels (17, 18). In human volunteers, peak serum levels following 7.5 mg/kg of body weight intramuscularly are 18 to 20 /xg/ml (19, 2 0 ) . Our studies in sick patients showed somewhat higher mean peak levels of 25.0 fig/ml. More importantly, amikacin is resistant to five of the seven bacterial enzymes that inactivate gentamicin and tobramycin, as well as one of the two enzymes that inactivate kanamycin. It has shown good in-vitro activity against many Gram-negative bacilli that are resistant to these aminoglycosides (5, 7 ) . With extensive use of gentamicin, these resistant forms are becoming more prevalent, and now account for 9% to 20% of all clinical isolates in several centers (2-4). The results of the present study indicate that amikacin is efficacious in the treatment of serious Gram-negative infections. In 24 of the 25 episodes of Gram-negative sepsis, there was a good clinical and bacteriological response to amikacin therapy. Six of seven patients with bacteremia responded satisfactorily. The one treatment failure occurred in a patient with septicemia who died within 40 hours of starting the drug. Four patients with urinary-tract infections had recurrence of bacteruria documented by routine urine cultures collected at 30 days. This is not surprising in view of the underlying structural de-
fects of their genitourinary tracts. There have been few published reports of the effectiveness of amikacin in serious infection. Sharp, Saenz, and Martin (21) showed that this agent successfully eradicated gentamicin-resistant Pr. rettgeri in 12 urinary-tract infections, including 1 with bacteremia. Our data, in addition to this report, suggest that amikacin will be useful in infections with gentamicin-resistant organisms. Previous animal studies have indicated that amikacin possesses otic and renal toxicity similar to kanamycin (22). We noted VHIth cranial nerve toxicity in two patients: one patient had three courses of amikacin therapy totalling 24 g for recurrent infections with a resistant Proteus, and the second patient had elevated trough amikacin blood levels. Thus, these patients had mitigating factors which may have contributed to the untoward reactions. Three patients developed an elevated serum creatinine while receiving amikacin. In two patients it was a transient event. The third patient had progressive azotemia and elevated serum amikacin levels following pseudomonas septicemia with hypotension. The signs of sepsis cleared, but serum creatinine continued to rise until he died from leukemia on the ninth hospital day. The contribution of amikacin to the elevated creatinine levels remains cryptic, because transient renal insufficiency is common in septic patients. There were no other adverse reactions noted in this study. The clinical effectiveness shown in this study holds the promise that amikacin will be useful in treating serious Gram-negative infections, particularly those involving gentamicin-resistant microorganisms. This enthusiasm is somewhat tempered by ototoxicity and nephrotoxicity, untoward reactions shared by other aminoglycosides. Future studies are needed to corroborate the clinical effectiveness and to further evaluate the toxic potential of this agent. ACKNOWLEDGMENTS: The authors thank Mrs. Nilda Jacobus for technical assistance. Grant support: from Bristol Laboratories, Division of BristolMeyers Co. Presented in part at the 14th Interscience Conference on Antimicrobial Agents and Chemotherapy, September 1974, San Francisco, California. Received 13 December 1974; revision accepted 23 May 1975. • Requests for reprints should be addressed to Francis P. Tally, M.D., Tufts-New England Medical Center, 171 Harrison Avenue, Boston, MA 02111. References 1. FINLAND M: Changing patterns of susceptibility of common bacterial pathogens to antimicrobial agents. Ann Intern Med 76:1009-1036, 1972 2. BRYAN LE, SHAHRABADI MS, VANDEN ELZEN HM: Gentamicin
resistance in pseudomonas aeruginosa: R-factor-mediated resistance. Antimicrob Agents Chemother 6:191-199, 1974 3. CHAD WICK P: Resistance of Pseudomonas aeruginosa to gentamicin. Can Med Assoc J 109:585-587, 1973 4. HOLMES RK, MINSHEW BH, GOULD K, et al: Resistance of
Pseudomonas aeruginosa to gentamicin and related aminiglycoside antibiotics. Antimicrob Agents Chemother 6:253-262, 1974 5. BENVENISTE R, DAVIES J: Mechanisms of antibiotic resistance in bacteria. Ann Rev Biochem 42:471-506, 1973 6. NAITO T, NAKAGAWA S, ABE Y, et al: Aminoglycoside antibiotics
II. Configurational and positional isomers of BB-K8. / Antibiot {Tokyo) 26:297-307, 1973 Tally etal. • Amikacin
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
487
7. PRICE KE, PURSIANO TA, DEFURIA MD, et al: Activity of
BB-K8 (Amikacin) against clinical isolates resistant to one or more aminoglycoside antibiotics. Antimicrob Agents Chemother 5:143-152, 1974 8. BODEY GP, STEWART D: In vitro studies of BB-K8, a new aminoglycoside antibiotic. Antimicrob Agents Chemother 4:186-192, 1973 9. PICKETT MJ, PEDERSEN MM: Nonfermentative bacilli associated with man: II. Detection and identification. Am J Clin Pathol 54:164-177, 1970 10. EDWARDS PR, EWING WH: Identification of Enterobacteriaceae,
3rd ed. Minneapolis, Burgess Publishing Co., 1972
15. ALCID DV, SELIGMAN SJ: Simplified assay for gentamicin in the presence of other antibiotics. Antimicrob Agents Chemother 3:559-561, 1973 16. Yu PKW, WASHINGTON JA II: Comparative in vitro activity of three aminoglycoside antibiotics: BB-K8, kanamycin and gentamicin. Antimicrob Agents Chemother 4:133-139, 1973 17. CLARKE
JT,
LIBKE
RD,
REGAMEY
C,
et
al:
Comparative
pharmacokinetics of amikacin and kanamycin. Clin Pharmacol Ther 15:610-616, 1974 18. YOUNG LS, HEWITT WL: Activity of five aminoglycoside antibiotics in vitro against Gram-negative bacilli and Staphylococcus aureus. Antimicrob Agents Chemother 4:617-625, 1973
11. BAUER AW, KIRBY WMM, SHERRIS JC, et al: Antibiotic sus-
19. BODEY GP, VALDIVIESO M, FELD R, et al: Pharmacology of
ceptibility testing by a standardized single disc method. Am J Clin Pathol 45:493-496, 1966
amikacin in humans. Antimicrob Agents Chemother 5:508-512, 1974
12. HARWICK HJ, WEISS P, FEKETY FR J R : Application of microti-
20. CABANA BE, TAGGART JG:
Comparative
pharmacokinetics
BB-K8 and kanamycin in dogs and humans. Antimicrob Chemother 3:478-483, 1973
in
Agents
tration techniques to bacteriostatic and bactericidal antibiotic susceptibility testing. / Lab Clin Med 72:511-516, 1968 13. CUTLER RE, BURTON MO: Correlation of serum creatinine concentration and kanamycin half-life. JAMA 209:539-542, 1969
21. SHARP PM, SAENZ CA, MARTIN RR: Amikacin (BB-K8) treat-
14. CHAN RA, BENNER EJ, HOEPRICH PD: Gentamicin therapy in
22. REIFFENSTEIN
ment of multiple-drug-resistant proteus infections. Agents Chemother 5:435-438, 1974
renal failure: a nomogram for dosage. Ann Intern Med 76:773778, 1972
488
October 1975 • Annals of Internal Medicine • Volume 83 • Number 4
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
JC,
HOLMES
SW,
HOTTENDORF
Antimicrob GH,
et
al:
Ototoxic studies with BB-K8, a new semisynthetic aminoglycoside. J Antibiot {Tokyo) 26:94-100, 1973
Amikacin (BB-K8) is a semisynthetic derivative of kanamycin which is active in vitro against many gentamicin-resistant Gram-negative bacilli. Twenty-three patients with 25 serious Gram-negative infections were treated with this new aminoglycoside. Twelve infections involved organisms that were resistant to gentamicin. Twenty patients satisfied the criteria for bacteriological and clinical cure. This included 11 of the 12 infections involving gentamicin-resistant Gram-negative bacilli. In 4 urinary tract infections there was a good clinical response, but routine follow-up urine cultures at 30 days were positive. One patient failed on amikacin therapy. Eighth nerve toxicity was detected in two patients. These results indicate that amikacin is effective in the treatment of serious Gram-negative infections and is particularly useful in those involving resistant organisms. Further studies are indicated to evaluate ototoxic potential.
have reported the emerging resistance of Enterobacteriaceae and Pseudomonads to gentamicin (Garamycin®*), a widely used antibiotic for serious Gram-negative infections (1-4). This changing pattern of antimicrobial susceptibility serves as a constant stimulus for the development of new antimicrobial agents and for the modification of existing drugs. The challenge is heightened by the existence of at least nine bacterial enzymes that are known to inactivate aminoglycoside antibiotics (5). Amikacin (BB-K8t) is a new semisynthetic derivative of kanamycin sulfate (Kantrex®t) obtained by acylation with an amino hydroxy butyric acid at the C-l amino group of the 2-deoxystreptamine moiety (Figure 1) (6). This new compound is a poor substrate for three of the four enzymes that inactivate gentamicin and for two of the three that inactivate tobramycin (Tobracin®t) (7). In-vitro susceptibility studies show that amikacin possesses good activity against Enterobacteriaceae and Pseudomonads, including many gentamicin-resistant microorganisms (8). The present study was undertaken to evaluate the SEVERAL LABORATORIES
clinical efficacy and toxicity of amikacin in the treatment of serious Gram-negative infections. Materials and Methods PATIENTS
The criterion for admission to the study was that all patients had signs of Gram-negative infection that indicated the need for antibiotic treatment and could be used as parameters of efficacy. These included fever, hypotension, chills, urinary-tract symptoms, and positive cultures. Additionally, an attempt was made to study infections involving resistant organisms according to the following guidelines: [1] previous isolation of Gramnegative pathogens that were resistant by in-vitro testing to conventional antimicrobials including penicillins, cephalosporins, and gentamicin; or [2] before the availability of bacteriological results, a clinical setting that suggested resistant bacterial pathogens. This category included patients with hospital-acquired infections and those with persistent sepsis despite several days of antimicrobial treatment with conventional agents. Informed consent was obtained from all patients or next-of-kin before amikacin treatment. MICROBIOLOGICAL STUDIES
Blood cultures were collected from all patients before, during, and after therapy. Patients with urinary-tract infections had quantitative urine cultures obtained before, during, and 30 days following therapy; bacterial isolates were considered pathogens if counts were greater than lOVml. Other bacterio-
* Schering Corporation, Bloomfield, New Jersey. t Bristol Laboratories, Syracuse, New York. t Eli Lilly & Co., Indianapolis, Indiana. • From the Infectious Disease Section, Veterans Administration Hospital, Sepulveda, California, and the UCLA School of Medicine, Los Angeles, California.
484
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
Figure 1 . Structural configuration of amikacin and kanamycin. Annals of Internal Medicine 83:484-488, 1975
logical studies included aspirates of closed spaces in appropriate situations. Nonfermentative Gram-negative bacilli were identified by the methods of Pickett and Pedersen (9). Other Gramnegative bacilli were identified according to Edwards and Ewing (10). Antimicrobial susceptibility of the isolates was initially tested by standardized disc-diffusion methods (11). Minimal inhibitory concentrations (MIC) and minimal bactericidal concentrations (MBC) were subsequently tested by broth dilution with microtiter techniques (12). The test medium was MuellerHinton broth; the inoculum was a 1-to-100 dilution of a 6-h broth culture to give a final concentration of approximately 103"4 colony forming U/ml. This inoculum was delivered to microtiter wells and the plates were incubated overnight in air at 37 °C. The MIC was the lowest concentration of antibiotic showing no macroscopic growth. The MBC was the lowest concentration showing 5 or less colonies on subculturing 0.05 ml from the microtiter wells to antibiotic free Mueller-Hinton agar.
Table 1 . Bacteriology of 25 Infections Treated with Amikacin
Organisms
Total Isolates
Blood Culture Isolates
Escherichia coli Pseudomonas aeruginosa Proteus rettgeri Serratia marcescens Proteus mirabilis Klebsiella species Providencia stuartii Proteus morganii Pseudomonas species Pseudomonas maltophilia Enterococcus Total
7 7 5 4 4 4 2 1 1 1 1 37*
3 2
1 1 1
8t
* Six patients had 2 isolates, 3 patients had 3 isolates. t One patient had 2 isolates.
DRUG DOSAGE
Amikacin therapy was initiated in doses of 500 mg every 12 hours. Patients with renal impairment received modified dosage according to established nomograms for aminoglycosides based on serum creatinine values (13, 14); blood levels of amikacin were used after initiating treatment to determine subsequent doses. In all cases, amikacin was administered as a single agent and previous antimicrobial agents were discontinued. SERUM LEVELS
Serum amikacin levels were measured by the agar well microbiological method using Staphylococcus epidermidis American Typed Culture Collection #27626 as the assay organism (15). Levels were obtained on the second day, and on the fifth or sixth day of therapy. Blood for peak values was drawn at 1 h and for troughs at 11 Vi h after intramuscular injection. MONITORING TOXICITY
Complete blood count with differential leukocyte count, urinalysis, serum creatinine, and liver function studies were obtained at least every 3 days during treatment. Audiometry was done whenever feasible before and after therapy with a Beltone Model 12-D portable audiometer* or I.A.C. 1204 series sound suitet. Results were kindly interpreted by Drs. D. Svihovec and W. Hanson of the Audiology and Speech Pathology Service of the Sepulveda Veterans Administration Hospital. PATIENT EVALUATION
Patients were considered cured if they became afebrile, all signs of infection cleared, and follow-up cultures showed eradication of the original pathogen. Patients who died subsequently of noninfectious diseases were considered cured if there was no evidence of the original infection at autopsy. Infections were judged drug failures if signs of infection persisted, follow-up cultures were positive, or alternative antimicrobial agents were required during the course of treatment. Results
Twenty-three patients with 25 infections fulfilled the criteria for admission to the study. All were men, reflecting the population of a Veterans Administration hospital. Twenty infections (18 patients) originated in the urinary tract; underlying conditions were obstructive uropathy in 9, neurologic disorders requiring suprapubic or indwelling Foley catheters in 7, nephrolithiasis in 3, and a sigmoid bladder following resection of a bladder carcinoma in 1. Three patients had septic arthritis; 2 had acute myelo* Beltone Electric Corporation, Chicago, Illinois. t Industrial Acoustics Co., Bronx, New York.
monocytic leukemia, and a third had a plastic knee prosthesis. One patient with chronic renal failure had an infection in the arteriovenous shunt. The final patient had an intraabdominal abscess following laparotomy. Fever was present at the onset of the study in 23 of the 25 infections; the mean peak rectal temperature was 39.4 °C. The 2 afebrile patients included 1 with septic arthritis, and a second with a urinary tract infection and the acute onset of hypotension. Additional signs of sepsis included leukocytosis (peripheral leukocyte count greater than 10 000/mm 3 ) in 20, hypotension (systolic blood pressure less than 90 mm Hg) in 9, and chills in 8. Bacteriologic studies in the 25 infections yielded a total of 37 pathogens; six patients had 2 isolates and three had 3 isolates from infected sites (Table 1). Escherichia coli and Pseudomonas aeruginosa were the most frequent organisms encountered. Proteus rettgeri, Serratia marcescens, Pr. mirabilis and Klebsiella were also common isolates. Seven patients had bacteremia, including one individual with both E. coli and Klebsiella in blood cultures. Bacteremic episodes occurred in 3 patients while they were receiving gentamicin before treatment with amikacin. Amikacin serum levels were obtained during treatment in all patients. The geometric mean peak level one h after injection was 25.0 /xg/ml with a range of 9.5 to 66 fig/ml. The high peak level was in a patient receiving hemodialysis. The mean geometric trough level was 3.5 /xg/ml with a range of 1.1 to 21 /xg/ml. The high trough level of 21 /xg/ml was in a patient with leukemia who had pseudomonas bacteremia, hypotension, and acute renal insufficiency. Based on mean achievable blood levels, microbial resistance was defined as an MIC greater than 16 /xg/ml for amikacin and greater than 8 tig/ml for gentamicin (16). Nine organisms were resistant to gentamicin (Table 2 ) . Four additional isolates (2 Pr. mirabilis, 1 Ps. aeruginosa, and 1 E. coli) had MIC's to gentamicin of 4 /xg/ml and were considered only intermediately susceptible. All but one of these resistant and marginally susceptible organisms were susceptible to amikacin. The exception was a Ps. maltophilia resistant to both gentamicin and amikacin that was recovered in a polymicrobial urinary tract infection. Tally et al. • Amikacin
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
485
Table 2. Resistant Organisms
Patient
Isolates
Source
Minimum Inhibitory Concentration Genta- Amikacin micin
culture of the joint fluid yielded Ps. aeruginosa susceptible to gentamicin. Despite adequate levels of gentamicin, purulent drainage continued and a gentamicin resistant isolate of Ps. aeruginosa was cultured. He was subsequently treated with amikacin, 500 mg, twice daily for 24 days. The drainage ceased, and the wound closed without removal of the prosthesis. His joint remains free of infection 1 year following treatment.
tig/ml
1 2 3 8 9 10 13 14 17 18 20 21 23
Pseudomonas aeruginosaKnee Aspirate Urine Serratia marcescens Proteus mirabilis Urine Pseudomonas aeruginosaBlood Escherichia coli Blood Proteus rettgeri Urine Proteus rettgeri Urine Urine Serratia marcesens Proteus mirabilis Urine Proteus rettgeri Urine Providencia stuartii Urine Pseudomonas maltophiliaUrine Serratia marcescens Urine
32 >64 4 4 4 32 36 64 4 64 64 >64 >64
PATIENT 7
0.5 4 16 1 8 16 16 8 4 4 16 >64 2
Five of the resistant isolates were susceptible to 128 fig/ ml or less of carbenicillin (Geopen®*) including 2 Pr. mirabilis, 1 Ps. aeruginosa, 1 Pr. rettgeri, and 1 Providencia stuartii. Disc-diffusion testing showed that the same 2 Pr. mirabilis were susceptible to cephalothin (Keflin®); another Pr. mirabilis and 1 Pr. rettgeri were inhibited by kanamycin. However, the other gentamicin-resistant isolates were resistant to these as well as other antibiotics. Evaluation of amikacin treatment in the 25 serious Gram-negative infections showed that 20 patients satisfied the criteria for clinical and bacteriological cure. Twelve of these infections involved bacteria that were resistant, or intermediately susceptible, to gentamicin; 6 had been treated with gentamicin and failed to respond—thus confirming the in-vitro data. In the other 6 cases, amikacin was used for initial chemotherapy. Eleven of the 12 gentamicin-resistant infections were cured with amikacin therapy. Three additional patients had persistent bacteremia during gentamicin therapy, although the bloodculture isolate was sensitive to gentamicin by in-vitro testing. In each instance the signs of sepsis cleared and blood cultures were negative following substitution of therapy with amikacin. Four infections were characterized by a good clinical response, but follow-up cultures were positive. All were urinary-tract infections in patients with underlying structural defects. In each instance, bacteruria was eradicated during treatment and in the 7-day follow-up urine culture. However, urine cultures at 30 days were again positive for the original organism. Illustrative Cases PATIENT 1.
A 55-year-old man with severe degenerative arthritis of the left knee underwent a total knee replacement with insertion of a plastic prothesis. He received prophylactic cephalothin during and following surgery. Postoperatively, he developed erythema, swelling, and pain at the incision. Sutures were removed, and * J. B. Roerig and Co., New York, New York. t Eli Lily & Co., Indianapolis, Indiana. 486
A 74-year-old man with parkinsonism and obstructive uropathy developed a fever of 40 °C. Both blood and urine cultures yielded Prov. stuartii that was resistant to gentamicin (MIC of 64 fig/ml). He was treated with amikacin, 500 mg, twice daily for 7 days. He defervesced in 36 h, and repeat blood and urine cultures were sterile. PATIENT 8
A 75-year-old man with acute myeloblasts leukemia developed fever to 40 °C, confusion, and hypotension. Laboratory data showed a leukocyte count of 1800/mm3 (17% neutrophils, 67% lymphocytes, 15% monocytes, and 1% eosinophils) and a platelet count of 16 760/mm3; the bone marrow showed many blast forms. Blood and urine cultures grew Ps. aeruginosa resistant to gentamicin. Following amikacin treatment, 500 mg, every 12 hours, he had a prompt defervescence with negative blood and urine cultures. During therapy the serum creatinine increased from 0.9 to 2.3 mg/100 ml. After 9 days of treatment, he had acute respiratory distress secondary to pulmonary edema and died. Autopsy showed acute myelocytic leukemia and pulmonary edema; there were no signs of residual infection. PATIENT 12
A 55-year-old man with renal failure on hemodialysis for 4 years developed swelling, erythema, and a purulent discharge at the arteriovenous shunt. Blood and wound cultures yielded Ps. aeruginosa susceptible to gentamicin. Despite 5 days of gentamicin treatment, he remained febrile with positive blood cultures. Therapy was changed to amikacin (3 doses of 500 mg during 72 hours) with rapid resolution of fever and clearing of the bacteremia. The infected shunt was then removed and antimicrobial treatment was discontinued. Blood levels of amikacin after the second dose showed a peak of 66 fig/vol with a trough of 4.5 fig/vol. Audiograms before therapy showed a mild highfrequency hearing loss, and there was no change in repeat studies 1 week following therapy. PATIENT 13
A 49-year-old man with advanced multiple sclerosis and a neurogenic bladder requiring suprapubic bladder drainage developed three separate episodes of Gram-negative sepsis. The first infection was characterized by flank pain, chills, fever of 40 °C, and a leukocytosis of 20 500 leukocytes/mm3. Urine culture yielded a gentamicin resistant Pr. rettgeri (MIC, 32 jitg/ml). Treatment with amikacin produced a prompt defervescence and clearing of bacteruria. However, urine culture at 30 days was again positive for the same organism. Two and 4 months later he developed similar signs of sepsis with a gentamicin-resistant Pr. rettgeri and Pr. mirabilis (MIC of 4 /ig/ml to gentamicin), respectively. On each occasion he had a prompt clinical response to amikacin. PATIENT 18
One patient failed on amikacin therapy. A 62-year-old man with atrial fibrillation and diabetes was admitted comatose with a left hemiparesis, retinal exudates, and papilledema. A Pr. mirabilis urinary tract infection was treated with ampicillin, 500 mg, intravenously every 6 h. Eight days later he developed a fever of 39.6 °C and leukocytosis of 35 100 leukocytes/mm3. At this time urine culture yielded a gentamicinresistant P. rettgeri and blood cultures were positive for E. coli; both strains were sensitive to amikacin. Amikacin therapy was initiated, but the patient continued to be febrile, and he died
October 1975 • Annals of Internal Medicine • Volume 83 • Number 4
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
40 hours later. Post-mortum examination showed a massive intracranial hemorrhage. The primary cause of death was attributed to the cerebral hemorrhage, but Gram-negative septicemia was considered a contributing factor. Pretreatment and posttreatment audiograms were available for 20 patients. Eighteen showed no change. However, one patient had a 15-dB threshold shift at 8 kHz after three courses of amikacin therapy totalling 24 g during a 6-month period. Another patient had a 20-dB shift in the midfrequency range. This individual received 7.25 g of amikacin during 10 days; the initial serum creatinine was 1.8 mg/100 ml, which decreased to 1.1 m g / 100 ml; but creatinine clearance remained unchanged (26 m l / m i n ) . Peak serum levels were only 27 and 21 ^g/ml, but trough levels were somewhat elevated to 12 and 7 fig/ml Renal function as measured by serum creatinine levels remained unchanged in 20 of 23 patients. One patient had an elevated serum creatinine associated with septic shock. He died of his underlying leukemia after 9 days of amikacin therapy. The cause of the renal insufficiency was thought to be secondary to an episode of hypotension, but the role of amikacin in this complication is unknown. Two additional patients with obstructive uropathy had a transient rise and fall in their serum creatinine levels during amikacin therapy. N o other complications were noted. Discussion
Amikacin (BB-K8) is a derivative of kanamycin with pharmacokinetics similar to those of the parent compound (17). It is active in vitro against many Enterobacteriaceae and Pseudomonads ( 7 ) . Comparative studies of various aminoglycosides have shown that amikacin is less active on a weight basis than gentamicin and tobramycin, but this disadvantage is counterbalanced by the higher achievable serum levels (17, 18). In human volunteers, peak serum levels following 7.5 mg/kg of body weight intramuscularly are 18 to 20 /xg/ml (19, 2 0 ) . Our studies in sick patients showed somewhat higher mean peak levels of 25.0 fig/ml. More importantly, amikacin is resistant to five of the seven bacterial enzymes that inactivate gentamicin and tobramycin, as well as one of the two enzymes that inactivate kanamycin. It has shown good in-vitro activity against many Gram-negative bacilli that are resistant to these aminoglycosides (5, 7 ) . With extensive use of gentamicin, these resistant forms are becoming more prevalent, and now account for 9% to 20% of all clinical isolates in several centers (2-4). The results of the present study indicate that amikacin is efficacious in the treatment of serious Gram-negative infections. In 24 of the 25 episodes of Gram-negative sepsis, there was a good clinical and bacteriological response to amikacin therapy. Six of seven patients with bacteremia responded satisfactorily. The one treatment failure occurred in a patient with septicemia who died within 40 hours of starting the drug. Four patients with urinary-tract infections had recurrence of bacteruria documented by routine urine cultures collected at 30 days. This is not surprising in view of the underlying structural de-
fects of their genitourinary tracts. There have been few published reports of the effectiveness of amikacin in serious infection. Sharp, Saenz, and Martin (21) showed that this agent successfully eradicated gentamicin-resistant Pr. rettgeri in 12 urinary-tract infections, including 1 with bacteremia. Our data, in addition to this report, suggest that amikacin will be useful in infections with gentamicin-resistant organisms. Previous animal studies have indicated that amikacin possesses otic and renal toxicity similar to kanamycin (22). We noted VHIth cranial nerve toxicity in two patients: one patient had three courses of amikacin therapy totalling 24 g for recurrent infections with a resistant Proteus, and the second patient had elevated trough amikacin blood levels. Thus, these patients had mitigating factors which may have contributed to the untoward reactions. Three patients developed an elevated serum creatinine while receiving amikacin. In two patients it was a transient event. The third patient had progressive azotemia and elevated serum amikacin levels following pseudomonas septicemia with hypotension. The signs of sepsis cleared, but serum creatinine continued to rise until he died from leukemia on the ninth hospital day. The contribution of amikacin to the elevated creatinine levels remains cryptic, because transient renal insufficiency is common in septic patients. There were no other adverse reactions noted in this study. The clinical effectiveness shown in this study holds the promise that amikacin will be useful in treating serious Gram-negative infections, particularly those involving gentamicin-resistant microorganisms. This enthusiasm is somewhat tempered by ototoxicity and nephrotoxicity, untoward reactions shared by other aminoglycosides. Future studies are needed to corroborate the clinical effectiveness and to further evaluate the toxic potential of this agent. ACKNOWLEDGMENTS: The authors thank Mrs. Nilda Jacobus for technical assistance. Grant support: from Bristol Laboratories, Division of BristolMeyers Co. Presented in part at the 14th Interscience Conference on Antimicrobial Agents and Chemotherapy, September 1974, San Francisco, California. Received 13 December 1974; revision accepted 23 May 1975. • Requests for reprints should be addressed to Francis P. Tally, M.D., Tufts-New England Medical Center, 171 Harrison Avenue, Boston, MA 02111. References 1. FINLAND M: Changing patterns of susceptibility of common bacterial pathogens to antimicrobial agents. Ann Intern Med 76:1009-1036, 1972 2. BRYAN LE, SHAHRABADI MS, VANDEN ELZEN HM: Gentamicin
resistance in pseudomonas aeruginosa: R-factor-mediated resistance. Antimicrob Agents Chemother 6:191-199, 1974 3. CHAD WICK P: Resistance of Pseudomonas aeruginosa to gentamicin. Can Med Assoc J 109:585-587, 1973 4. HOLMES RK, MINSHEW BH, GOULD K, et al: Resistance of
Pseudomonas aeruginosa to gentamicin and related aminiglycoside antibiotics. Antimicrob Agents Chemother 6:253-262, 1974 5. BENVENISTE R, DAVIES J: Mechanisms of antibiotic resistance in bacteria. Ann Rev Biochem 42:471-506, 1973 6. NAITO T, NAKAGAWA S, ABE Y, et al: Aminoglycoside antibiotics
II. Configurational and positional isomers of BB-K8. / Antibiot {Tokyo) 26:297-307, 1973 Tally etal. • Amikacin
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
487
7. PRICE KE, PURSIANO TA, DEFURIA MD, et al: Activity of
BB-K8 (Amikacin) against clinical isolates resistant to one or more aminoglycoside antibiotics. Antimicrob Agents Chemother 5:143-152, 1974 8. BODEY GP, STEWART D: In vitro studies of BB-K8, a new aminoglycoside antibiotic. Antimicrob Agents Chemother 4:186-192, 1973 9. PICKETT MJ, PEDERSEN MM: Nonfermentative bacilli associated with man: II. Detection and identification. Am J Clin Pathol 54:164-177, 1970 10. EDWARDS PR, EWING WH: Identification of Enterobacteriaceae,
3rd ed. Minneapolis, Burgess Publishing Co., 1972
15. ALCID DV, SELIGMAN SJ: Simplified assay for gentamicin in the presence of other antibiotics. Antimicrob Agents Chemother 3:559-561, 1973 16. Yu PKW, WASHINGTON JA II: Comparative in vitro activity of three aminoglycoside antibiotics: BB-K8, kanamycin and gentamicin. Antimicrob Agents Chemother 4:133-139, 1973 17. CLARKE
JT,
LIBKE
RD,
REGAMEY
C,
et
al:
Comparative
pharmacokinetics of amikacin and kanamycin. Clin Pharmacol Ther 15:610-616, 1974 18. YOUNG LS, HEWITT WL: Activity of five aminoglycoside antibiotics in vitro against Gram-negative bacilli and Staphylococcus aureus. Antimicrob Agents Chemother 4:617-625, 1973
11. BAUER AW, KIRBY WMM, SHERRIS JC, et al: Antibiotic sus-
19. BODEY GP, VALDIVIESO M, FELD R, et al: Pharmacology of
ceptibility testing by a standardized single disc method. Am J Clin Pathol 45:493-496, 1966
amikacin in humans. Antimicrob Agents Chemother 5:508-512, 1974
12. HARWICK HJ, WEISS P, FEKETY FR J R : Application of microti-
20. CABANA BE, TAGGART JG:
Comparative
pharmacokinetics
BB-K8 and kanamycin in dogs and humans. Antimicrob Chemother 3:478-483, 1973
in
Agents
tration techniques to bacteriostatic and bactericidal antibiotic susceptibility testing. / Lab Clin Med 72:511-516, 1968 13. CUTLER RE, BURTON MO: Correlation of serum creatinine concentration and kanamycin half-life. JAMA 209:539-542, 1969
21. SHARP PM, SAENZ CA, MARTIN RR: Amikacin (BB-K8) treat-
14. CHAN RA, BENNER EJ, HOEPRICH PD: Gentamicin therapy in
22. REIFFENSTEIN
ment of multiple-drug-resistant proteus infections. Agents Chemother 5:435-438, 1974
renal failure: a nomogram for dosage. Ann Intern Med 76:773778, 1972
488
October 1975 • Annals of Internal Medicine • Volume 83 • Number 4
Downloaded From: https://annals.org/ by a Univ of Utah User on 09/25/2017
JC,
HOLMES
SW,
HOTTENDORF
Antimicrob GH,
et
al:
Ototoxic studies with BB-K8, a new semisynthetic aminoglycoside. J Antibiot {Tokyo) 26:94-100, 1973
