Intravenous Gentamicin Therapy for Infections in Patients with Cancer

Transcription

1 TH JOURNAL OF INFCTIOUS DISASS VOL. 124, SUPPLMNT DCMBR by the University of Chicago. All rights reserved. Intravenos Gentamicin Therapy for Infections in Patients with Cancer Gerald P. Bodey,* dward Middleman, Theera Umsawasdi, and Victorio Rodrigez From the Department of Developmental Therapetics, the University of Texas M.D. Anderson Hospital and Tmor Institte at Hoston, and the Department of Medicine, Brooke General Hospital, Fort Sam Hoston, Texas Gentamicin slfate was administered intravenosly to 60 patients dring 75 episodes of infection. A dose of 30 mg of gentamicin/rn'' was given every 6 hr, after an initial dose of 40 mg/rn-. Forty-one of the infections (55%) responded favorably. Only 32% of the infections in patients with an initial netrophil cont of less than 100/mm 3 responded favorably, as compared to 60% of the infections in patients with an initial netrophil cont of more than 1,000/mm 3. Azotemia occrred in 21 of the 57 patients who had normal initial levels of blood rea nitrogen and who received gentamicin for at least seven days. Gentamicin appears to be an effective antibiotic in cancer patients nless they have severe netropenia. The drg can be administered safely by the intravenos rote. Since effective therapy is now available for some malignant diseases, sccessfl management of infections in patients with cancer has become increasingly important. Althogh the cancer patient is ssceptible to a wide variety of organisms, the majority of infections are cased by gramnegative bacilli. Dring the three-year period from 1967 to 1969, 85% of the 493 episodes of bacteremia observed at or instittion were cased by gram-negative bacilli [1]. The organisms cltred most freqently from specimens of blood were scherichia coli, Psedomonas sp., and Klebsiella sp. A major advantage of gentamicin slfate over most other antibiotics is its activity against both nterobacteriaceae and species of Psedomonas. Frthermore, an increasing nmber of infections are being cased by organisms that are sensitive only to this antibiotic. At or instittion a nonpigmented strain of Serratia sp., which is consistently sensitive only to gentamicin, has cased a nmber of serios infections [2]. Infectios complications are especially freqent in patients with lekemia. Forty-five episodes of Please address reqests for reprints to Dr. Gerald P. Bodey, M.D. Anderson Hospital and Tmor Institte, Texas Medical Center, Hoston, Texas * Dr. Bodey is a scholar of the Lekemia Society of America, Inc. septicemia occrred per 100 admissions among patients with lekemia, compared to 2.4 episodes among patients with metastatic carcinoma [3]. Lekemic patients generally have thrombocytopenia, and it is hazardos for them to receive medications by the intramsclar rote. Conseqently, stdies of gentamicin slfate administered iv were initiated. These stdies are also applicable to patients in shock, since the absorption of medications from mscle is compromised nder these circmstances. Methods Gentamicin slfate was administered iv to 60 patients dring 75 episodes of infections. Fifty-six infections occrred in patients with lekemia, for in patients with lymphoma, and 15 in patients with metastatic carcinoma. The median age of the patients was 44 years old (range of 6-76 years of age). Dring 49 infections, the patients initially received other antibiotics (sally carbenicillin or cephalothin). These antibiotics were contined after gentamicin was added dring 27 infections. However, in every instance the previos antibiotic therapy was proven to be ineffective before gentamicin was administered. For additional patients received cephalothin pls gentamicin as initial therapy for infections de to Psedomonas. Patients were considered to have had a com- 8174

2 Gentamicin for Infections in Cancer Patients S175 plete response if they became afebrile, and all signs of infection disappeared. Patients who died of sperinfection or of their nderlying malignancy were considered to have had a complete response if their original infection resolved. Three patients were considered to have had a partial response becase they had objective and symptomatic improvement, bt their infection was not completely eradicated. Gentamicin slfate was administered iv at a dose of 30 mg/m" (0.75 rg/kg) every 6 hr, after an initial dose of 40 mg/m" (1 rg/kg). The drg was dissolved in 200 ml of 5% glcose soltion and infsed over a 2-hr period. The schedle of dosage was modified for patients with renal impairment [4]. Ssceptibility tests were condcted on 413 clinical isolates of gram-negative bacilli, sing the tbe-diltion techniqe [5]. The organisms were inoclated into Meller-Hinton broth (Difco) and incbated at 37 C for 18 hr. A O.l-ml aliqot of a 10-4 diltion of this broth (approximately 10 3 cf) was sed as the inoclm for ssceptibility testing. Gentamicin slfate was dissolved in Meller-Hinton broth to a final concentration of 100 g/ml, Twofold serial diltions were made with Meller-Hinton broth, and the MIC was determined after incbation at 37 C for 18 hr. Reslts Figre 1 shows the in-vitro ssceptibility to gentamicin slfate of 413 isolates of gram-negative bacilli cltred from specimens of blood obtained from cancer patients. xclding Protes, the MIC of gentamicin for over 95% of these isolates was 1.56 g/rnl or less. All of the isolates of Protes sp. were inhibited by 3.12 g/rnl or less. Forty-one of the 75 episodes of infection (55%) responded to gentamicin therapy (table 1). The drg was most effective for the treatment of celllitis and rinary-tract infection. None of the seven episodes involving mltiple sites of infection responded. Thirty-seven (41 %) of the infections were associated with bacteremia, and 19 (48 %) of these infections responded favorably. Psedomonas aerginosa was the etiologic agent isolated most freqently from these patients (table 2). Forteen episodes of infection were cased by mltiple organisms, and no etiologic Q... -ן m In...!;i..J o IL o ffi MINIMAL INHIBITORY CONCNTRATION (IIQ I ml) Figre 1. In-vitro ssceptibility of 413 clinical isolates of gram-negative bacilli to gentamicin slfate. agent cold be isolated from the cltre specimens obtained from the patients dring nine episodes. Twenty-eight (57 %) of the 49 episodes of infection cased by a single species of gram-negative bacilli responded to therapy with gentamicin. Infection de to Protes sp. and Klebsiella-nterobacter had the highest rates of favorable responses. Only seven (47 %) of the 15 infections de to Psedomonas responded to gentamicin therapy. Table 1. Response to intravenos gentamicin therapy by site of infection. Site Pnemonia Celllitis Urinary-tract infection Bacteremia alone Mltiple sites Other All bacteremia Total Total no % Response o

3 8176 Bodey et al. Table 2. Response to intravenos gentamicin therapy by infecting organism. Organism Total no. % Responses Psedomonas Serratia Klebsiella-nterobacter scherichia coli 5 80 Protes Other gram-negative bacilli 4 50 Gram-positive cocci 3 67 Mltiple organisms Unknown 9 56 Total The three partial responses were a rinary-tract infection de to Protes mirabilis which relapsed when gentamicin was discontined, a lng abscess de to Serratia sp. which reqired srgical drainage, and a pelvic abscess cased by mltiple organisms in a patient with extensive tmor that recrred when gentamicin was discontined. The effectiveness of gentamicin was related to the patients' level of circlating netrophils (table 3). Only 32% of the infections in patients with an initial netrophil cont of less than 100/mm s responded, as compared to 60% of the infections in patients with an initial netrophil cont of more than 1,000/mm 3 However, the latter grop inclded the majority of infections cased by mltiple organisms that occrred in patients following srgery of the gastrointestinal or genitorinary tracts; these infections sally respond poorly to all antibiotic therapy. The relationship between netrophil cont and response to gentamicin therapy is more obvios when only infections cased by single species of gram-negative bacills are considered. Five (31 % ) of the 16 infections in patients with less than 100 netrophils/rnm" responded favorably compared to 10 (83 %) of the 12 infections in patients with more than 1,000 netrophils/mm", The case smmarized in figre 2 illstrates the Table 3. Response to intravenos gentamicin therapy related to netrophil cont. Gram-negative Initial Total infections bacilli infections netrophil Total % Total % cont/rnmf no. Response no. Response < > relation of the level of circlating netrophils to the response to gentamicin therapy. The patient, a 49-year-old women, developed severe netropenia while receiving therapy for acte myelogenos lekemia. Carbenicillin and cephalothin were administered initially when she became febrile. Gentamicin was sbstitted for carbenicillin when Klebsiella sp. was cltred from two specimens of blood. She remained febrile for 12 days while receiving this antibiotic regimen, and dring this time she had no circlating netrophils. Blood cltres were negative dring this time, bt Klebsiella sp. was isolated from two blood cltres after gentamicin therapy was discontined. Gentamicin was reinstitted and cephalothin was discontined. Simltaneosly the patient achieved a remission of her lekemia, and the netrophil cont retrned to normal levels. Her infection responded promptly to gentamicin therapy when her netrophil cont retrned toward normal. The ineffectiveness of gentamicin in patients with persistent netropenia is illstrated by the corse of a 19-year-old man with acte lymphocytic lekemia (figre 3). While ndergoing chemotherapy he developed severe netropenia and became febrile. On the sixth day of fever, Serratia sp. was cltred from a specimen of blood, and therapy was institted with 120 mg of gentamicin/m" per day and ampicillin. On day 16 he developed areas of celllitis on the chest and forehead, and Serratia sp. was cltred from a specimen of blood drawn on day 21. The organism was resistant in vitro to all antibiotics except gentamicin. The dosage of gentamicin was escalated to 160 mg/m" per day and then to 300 mg/m" per day, reslting in moderate vestiblar and aditory toxicity. At this time both Serratia sp. and scherichia coli were grown from blood cltres. Chloramphenicol was sbstitted for ampicillin, bt the patient died of his infection. The MIC of gentamicin for the strain of Serratia sp. obtained from the first blood cltre was 0.78 g/ml and that of the sbseqent isolate was 1.56 g/rnl (figre 4). The MIC for the initial isolate of. coli was 1.56 g/ml, bt that of later isolates was 6.25 g/ml. On day 25, while the patient was receiving 40 mg/m" (75 mg), gentamicin every six hors, specimens of serm were obtained at the onset of an infsion of gentamicin and 1 hr

4 Gentamicin for Infections in Cancer Patients SI W 0: ffi a.. w f t fr--+--t l---l.---- l 102 -r-tt--tt--hr--tt tt---rit_+_+--r-- I+I---- l 100 -t--:-t-t-t \f+t---t+--\-i-h't--r<:f-.lf--y---l l I.--J L--.--l...L- ---l.-_----i...,..l-_j KLBSILLA SPTICMIA t-====== ::i " en...j Ld t _: :,L,..L--...j II 13 DAY OF INFCTION Figre 2. Clinical corse of a 49-year-old female with acte myelogenos lekemia and severe netropenia who developed septicemia de to Klebsiella sp, She only responded to therapy with gentamicin when her netrophil cont recovered. The arrows indicate when Klebsiella sp. was cltred from blood. 105 w 104 a:: i= 103 w w I en :j 0 w SRRATIA CLLULITIS AND SPTICMIA 15.COLI SPTICMIA WHI.aa:caUNT. A --==-C GRANULOCYTS DAY Figre 3. Clinical corse of a 19-year-old male with acte lymphocytic lekemia and severe netropenia who developed septicemia de to Serratia sp. and scherichia coli. The patient failed to respond to gentamicin therapy and died of his infection. The arrows indicate when the organisms were cltred from blood; S =Serratia, =. coli. later. Twofold diltions of these specimens of serm were tested in vitro against the isolates of Serratia sp. and. coli cltred from the patient's blood. All of the organisms were inhibited by his serm in vitro at diltions of 1:4 or greater. The toxic effects of gentamicin were ascertained dring these 75 episodes of infection and 32 additional episodes of infection occrring in 23 patients who received gentamicin and carbenicillin [6]. One patient developed an erythematos

5 S178 Bodey et al. g 625.., 3.12 >..,...J 1.56 c ' 'e , c: (!) 0.39 MINIMUM INHIBITORY CONCNTRATiON Table 4. Nephrotoxicity associated with intravenos gentamicin therapy. Blood rea nitrogen Patients Initial Sbseqent No. Percent Normal Normal Normal levated levated Unchanged or lower 5 56 levated Frther elevated 4 44 c 32 /6 8 zs (J) 4 2 S S SRUM INHIBITORY ACTIVITY s s s DAY Figre 4. Reslts of in-vitro stdies with scherichia and Serratia sp. cltred from specimens of blood from a 19-year-old male with acte lymphocytic lekemia and severe netropenia. The pper portion of the figre shows the ssceptibility of the organisms to gentamicin slfate. The lower portion shows the inhibitory activity of twofold diltions of the patient's serm against the organisms cltred from his blood on the days indicated. The serm was obtained on day 25 when the patient was receiving 40 mg of gentamicin/m'' (75 mg) every 6 hr. Blood specimens were collected at the beginning (illj) of an iv infsion of gentamicin, (4 hr after the completion of the preceding infsion) and 1 hr later O). S = Serratia, =. coli. rash de to gentamicin, and three patients developed bllos lesions at sites where gentamicin infiltrated into the skin, which sbseqently became necrotic. Aditory toxicity occrred dring for of 107 corses of therapy. Three of the patients with ototoxicity also had azotemia; only one patient, who had diminished hearing before treatment started, became permanently deaf. The forth patient also received ethacrynic acid, which was felt to have contribted to his deafness. Serial determinations of blood rea nitrogen (BUN) were obtained from 66 patients who received gentamicin for at least seven days (table 4). Of the 57 patients who had normal vales initially, 21 developed azotemia. Three additional patients developed azotemia after receiving gentamicin for less than seven days. The median highest BUN was 34 mg/ioo ml (range, mg/ioo ml). Six patients developed BUN levels exceeding 50 mg/ioo ml. For of the nine patients with preexisting azotemia had frther progression dring gentamicin therapy. Discssion Gentamicin was effective against 55% of infections occrring in cancer patients. It had broadspectrm activity against infections cased by nterobacteriaceae and Psedomonas sp. It was least effective against infections involving mltiple sites or cased by mltiple organisms. Most of these patients had received other antibiotics before gentamicin was administered. It was apparent in every instance that the previos antibiotic therapy was ineffective. Dring 27 infections, the prior antibiotic was contined when gentamicin was initiated and, additionally, for patients received cephalothin pls gentamicin as initial therapy of infections de to Psedomonas sp. It is nlikely that this enhanced the effect of gentamicin since only 11 of these 31 infections responded (35%) compared to 30 of the 44 infections (68 %) where gentamicin was administered alone. Unfortnately, gentamicin is only minimally effective against infections in patients with netropenia. Netropenia is common in patients ndergoing chemotherapy for cancer, and especially in patients with acte lekemia. Infections are very freqent in these patients and are sally cased by gram-negative bacilli, especially Psedomonas sp. [7, 8]. The polymyxin antibiotics also are only minimally effective in patients with netropenia, whereas the penicillins are effective regardless of the patients' netrophil conts [8, 9]. Levels of gentamicin in serm have been determined in five patients with normal creatinine clearances who received gentamicin iv [10]. On

6 Gentamicin for Infections in Cancer Patients S179 the third day of therapy, the mean initial serm level (4 hr after completion of the preceding infsion) was 1.1 g/rnl. The mean peak level was 4.2flg/ml and was obtained at 2 hr, with a serm half-life of approximately 2 hr. These reslts were similar to those obtained following the im administration of gentamicin [11]. We have also stdied the combination of gentamicin and carbenicillin in the treatment of 32 infections in 23 cancer patients [6]. ighteen (56%) of these 32 infections and 12 (57%) of the 21 infections cased by single organisms responded to this combination. These reslts were the same as those obtained in the present stdy sing gentamicin alone. The combination was effective against five of six infections in which no etiologic agent cold be identified, whereas gentamicin alone was effective against five of nine infections. The patients had netrophil conts greater than I,OOO/mm 3 dring 25% of the infections treated with carbenicillin and gentamicin, compared to 38% of the infections treated with gentamicin alone. A similar relationship was observed between the netrophil cont and response to therapy with the combination for those infections cased by organisms resistant to carbenicillin. All of the infections de to Protes sp. and Psedomonas sp. responded to the combination, regardless of the patient's netrophil cont. This was probably de to the fact that carbenicillin is effective even in netropenic patients. Gentamicin, like the other aminoglycoside antibiotics, has oto- and nephrotoxicity. Many of or patients had received other nephrotoxic antibiotics before receiving gentamicin and were seriosly ill. This probably potentiated the nephrotoxicity of gentamicin. The iv administration of gentamicin was not associated with any niqe toxicity. Gentamicin is an antibiotic with broad-spectrm activity against gram-negative bacilli which is effective in the therapy of infections even in patients with widespread metastatic carcinoma. However, it is of limited benefit when sed alone in patients with severe netropenia. The drg can be given safely by the iv rote, and this method of administration shold be sed in patients who are severely ill, in shock, or have thrombocytopenia. References 1. Bodey, G. P. Spportive care of the cancer patient. Postgrad. Med. 48: , Bodey, G. P., Rodrigez, V., Smith, J. P. Serratia species infections in cancer patients. Cancer 25: , Bodey, G. P. Infections and cancer: natral history, diagnosis and treatment. In Oncology, 1970, diagnosis and management of cancer, general considerations (Proceedings of the 10th international cancer congress). Yearbook, Chicago (in press). 4. Gingell, J. C., Chisholm, G. D., Calnan, J. S., Waterworth, P. M. The dose, distribtion, and excretion of gentamicin with special reference to renal failre. J. Infect. Dis. 119: , Grove, D. A., Randal, W. A. Assay methods of antibiotics: a laboratory manal. Medical ncyclopedia, New York, 1967, p Rodrigez, V., Whitecar, J. P., Jr., Bodey, G. P. Therapy of infections with the combination of carbenicillin and gentamicin. Antimicrobial Agents and Chemotherapy-1969, p , Bodey, G. P., Bckley, M., Sathe, Y. S., Freireich,. J. Qantitative relationships between circlating lekocytes and infection in patients with acte lekemia. Ann. Intern. Med. 64: , Whitecar, J. P., r-; Bodey, G. P., Lna, M. Psedomonas bacteremia in patients with malignant diseases. Amer. 1. Med. Sci. 260: , Bodey, G. P., Rodrigez, V., Whitecar, 1. P. Severe infections in lekemic patients-an approach to antibiotic therapy. In F. Hoffman [ed.] Advances in the management of psedomonas and protes infections. Proceedings of the symposim, Jly 12, xerpta Medica, New York, 1970, p Rodrigez, V., Stewart, D., Bodey, G. P. Gentamicin slfate distribtion in body flids. Clin. Pharmacol. Thee. 11: , Black, J., Calesnick, B., Williams, D. Weinstein, M. J. Pharmacology of gentamicin, a new broadspectrm antibiotic. Antimicrobial Agents and Chemotherapy-1963, p , 1964.