266

Changing pattern of antibiotic sensitivityof Salmonella typhi

C. RANJU, P. PAIS, G.D. RAVINDRAN, G. SINGH

ABSTRACTBackground. The emergence of multidrug-resistant Salmo-

nella typhi led to the use of quinolones as the first-line drug in thetreat-ment of adult patients with typhoid fever. However, overthe last few years there has been an impression that patients onciprofloxacin tended to take longer to defervesce. We studied theresponse and antibiotic sensitivity patterns during 2 time periodsto assess the changes that may have occurred.

Methods. A retrospective analysis was done of blood culture-positive patients with Salmonella typhi infection during 1991 and1996-97. The mode of presentation, treatment history,antibiotic sensitivity pattern, antibiotics administered, responseto therapy and the complications that ensued were studied.

Results. in vitro sensitivity to ciprofloxacin was found to be100% in both the study groups. It was found that a greaternumber of patients were sensitive to ampicillin (80%), chloram-phenicol (80%) and co-trimoxazole (80%) during 1996-97 ascompared to 1991, when sensitivity to ampicillin was 63%,chloramphenicol 65% and co-trimoxazole 65%. The mean(SD) defervescence period in 1991 was 6 (2.3) days and in1996-97 was 6 (2) days (p>O.OS).

Conclusion. In vitro sensitivity of Salmonella typhi to clpro-floxadn remains 100%. There was an increase in the sensitivityto ampicillin, chloramphenicol and co-trimoxazole which havebeen rarely used over the past few years. There was no significantdifference in the time taken to defervesce between the two studyperiods.Natl Med J India 1998; 11:266-7

INTRODUCTIONImproved standards of public health hav~ resulted in a markeddecline in the incidence of typhoid fever in developed countries.'However, the disease continues to be rampant in Third Worldcountries.' The emergence of strains of Salmonella typhi (S. typhi)resistant to multiple antibiotics poses a serious problem. Resis-tance to chloramphenicol which has been in use since 1948,2wasfirst reported in 1950.3 Gradually, resistance to multiple antibio-tics developed+" and the first major epidemic of multidrug-resistant S. typhi was reported in 1972.IO Since then, an increasingfrequency of antibiotic resistance has been reported from all partsof the world, but more so from developing countries," Over the last5 years, quinolones have become the first line of treatment in themanagement of adult patients with S. typhi infection in ourhospital. The use of chloramphenicol, ampicillin and co-trimoxa-zole has become infrequent. There was an impression that patients

St John's MedicalCollegeand Hospital,Bangalore560034, Karnataka,India

C. RANJU,P. PAIS, G. D. RAVINDRAN,G. SINGHDepartmentof Medicine

Correspondenceto C. RANJU

© The National Medical Journal of India 1998

THENATIONALMEDICALJOURNALOF INDIA VOL. 11, No.6, 1998

treated with ciprofloxacin took longer to defervesce. Therefore,we studied the sensitivity pattern of S. typhi to chlorampheni-col, ampicillin and co-trimoxazole which have been infrequentlyused in the last 5 years.

PATIENTS AND METHODSA retrospective analysis was done of 100 consecutive patientswhose blood cultures were positive for S. typhi in 1991 and 1996-97. They were identified from the records of the microbiologydepartment. The sensitivity pattern of the blood cultures wasrecorded. Blood cultures had been prepared by inoculating 3 ml ofblood in 30 ml of broth and the sensitivity tested using the standardKirby Bauer method. The minimum inhibitory concentration (MIC)of the antibiotics had been determined by the tube dilution methodusing nutrient broth. The clinical course, laboratory tests andantibiotics used were recorded from the patient's case records.Defervescence was defined as the number of days required forabatement of fever after starting antibiotics.

Statistical analysisStudent t-test was used for continuous variables and the Chi-square test for categorical variables. A p value of <0.05 wasconsidered significant.

RESULTSThe mean (SD) age of the patients at presentation was 24 (11)years in both the study groups. Men predominated in both thestudy groups, 65% and 72% during 1991 and 1996-97, respec-tively. However, this difference was not statistically significant(p=0.29) The distribution of symptoms and signs in both groupswas similar with fever being the most common presentingfeature. The mean (SD) duration of fever prior to hospitalizationwas 10 (7) days and 9 (7) days during 1991 and 1996-97,respectively (p>0.05). The spleen was palpable in 44% and 42%and the liver in 23% and 18% in 1991 and 1996-97, respectively.Other less common features such as dysuria, arthralgia andaltered sensorium were seen in 10% and 13% of patients in 1991and 1996-97, respectively. Relative bradycardia was present in20% and 33% of the patients in 1991 and 1996, respectively.

Leucopenia was seen in 20% and 24% of patients in 1991 and1996-97, respectively and a single estimation of Widal wassuggestive of enteric fever in significant titres in 35% of thepatients only (0 titre of 1:160 or more).

Antibiotic sensitivity patternTable I shows that the sensitivity of S. typhi to ampicillin,chloramphenicol and co-trimoxazole was significantly higher in1996-97 than in 1991. However, cephalosporin sensitivity haddecreased from 99% to 88% (p<0.05), though the sensitivity ofciprofloxacin was 100% at both times.

Pattern of drug usageTable II shows the pattern of antibiotic administration. Cipro-

TABLEI. Antibiotic sensitivity pattern during 1991 and 1996-97

Antibiotic 1991 1996-97 p value

Ampicillin 63 80 0.007Chloramphenicol 65 80 0.017Co-trimoxazole 65 80 0.017Ciprofloxacin 100 100Norfloxadn 100 100Cephalosporins 99 88 0.001

RANJU et al. : ANTmIOTIC SENSITIVITY OF Salmonella typhi

TABLE II. Drug administration pattern

Antibiotic 1991 1996-97Initial Switch Second-line Initial Switch Second-line

Ampicillin 4 2 2 4 0 0Chloramphenicol 18 12 0 11 2 0Cephalosporin 4 I I 12 I 4Ciprofloxacin 49 0 23 45 0 5Co-trimoxazole 15 12 0 16 0 0Others' 10 I 2 12 6 0• aminoglycosides, amoxycillin, azithromycin

floxacin was the most commonly used antibiotic during both thestudy periods, as the initial or second-line antibiotic. In 1996,only 2 patients on chloramphenicol required to be shifted to asecond-line antibiotic while none of the patients on ampicillinand co-trimoxazole required a change in antibiotic therapy. Thiswas in contrast to the situation in 1991 when 2,12 and 12 patientson ampicillin, chloramphenicol and co-trimoxazole, respectively,were changed to other antibiotics due to non-response.

Defervescence periodThis ranged from 2 to 16 days in both the groups with a mean (SD)of 6 (2.3) days in 1991 and 6 (2) days in 1996. The defervescenceperiod with ciprofloxacin was 6.4 (2.4) and 6.1 (2) days, ampicillin5.5 (3) and 3.5 (1.7) days, and with co-trimoxazole 5.2 (1.7) and 5.3(1.2) days, respectively in 1991 and 1996. None of these differenceswere significant.

ComplicationsIn 1991, 10 patients developed complications; 3 had encephalopa-thy, 3 hepatitis, 2 gastrointestinal bleeding and 1 each developedagranulocytosis and acute renal failure. During 1996, 11 patientsdeveloped complications; 2 had encephalopathy, 2 gastrointesti-nal bleeding, 4 hepatitis, 2 pleural effusion and 1 developedpneumonia. The case-fatality rate during 1991 and 1996 was 2% and1%, respectively.

DISCUSSIONThere were no significant differences in the pattern of clinicalpresentation of typhoid fever in the two time periods. Thepredominant presenting feature was fever. Similar figures havebeen reported from Florida and Thailand. IO Neurological andupper gastrointestinal bleeding were quite rare. Leukopeniawas seen in only 20% of the cases. A single Widal estimationgave a diagnostic titre (0)1:160) in only 35% of the cases,suggesting that this test was not sensitive. This finding hasbeen supported by other studies also.'>!'

In the mid-1980s, increasing resistance of S. typhi to chloram-phenicol, co-trimoxazole and ampici1lin/amoxycillin (drugs whichwere in use at that time) was reported in many studies.v? Agarwalet al. reported chloramphenicol resistance of 2.7%, 5.3% and 26.5%andampicillin resistanceof5.4%, 28.2% and 24.6%in 1973, 1978and1979, respectively." Shridhar et al. in a 5-year study conducted inBangalore reported chloramphenicol resistance of 1.7%, 22.2%and 43.4% and ampicillin resistance of 24.5%, 14.4% and 11.7% in1977, 1979 and 1980, respectively." In the 1990s, multidrug resis-tance was reported more frequently (61.4%-75%).11.18 As a result,quinolones were used as front-line drugs in non-pregnant adultswith typhoid. Our findings suggest that while the sensitivity of S.typhi to quinolones has remained the same, there is an increase inits sensitivity to ampicillin, chloramphenicol and co-trimoxazole.This is also supported by a decrease in the need to change

267

antibiotics in 1996 as compared to 1991 (Table Il), indicating thatthe in vivo resistance to these antibiotics has also decreased. Wealso found development of resistance to cephalosporins, althoughthis effect was not observed in clinical practice. Ceftriaxone hasbeen used as a second-line drug in patients showing unsatisfac-tory response to quinolones and the development of resistance toit is cause for worry.

Clinicians have increasingly felt that the period of deferves-cence in patients treated with quinolones has increased overthe years. Our study does not support this. However, a strainof S. typhi has been reported from Vellore which thoughsensitive to ciprofloxacin, requires higher minimum inhibitoryconcentrations (MIC) to contain its growth." We have notcompared the MIC of the organisms in 1991 and 1996-97.

Our data suggest that the sensitivity of S. typhi to quinolonesremains complete. There has been a small but significant de-crease in the in vitro sensitivity to cephalosporins. This couldthreaten the utility of drugs such as ceftriaxone for the manage-ment of patients with complications, prolonged fever or whenquinolones are contraindicated. However, the sensitivity tochloramphenicol, ampicillin and co-trimoxazole has increasedover the past 5 years suggesting that these drugs could be usedagain in the management of typhoid fever. A prospective studyto answer this question needs to be performed.

ACKNOWLEDGEMENTWe are grateful to Drs Rajani Macaden, Prabha Unnikrishnan,Muralidharan S, Department of Microbiology, St. John's Medical Col-lege and Hospital, Bangalore for their assistance.

REFERENCESGulati PO, Saxena SN, Gupta PS. Changing pattern of typhoid fever. Am J Med1968;45:544-8.

2 Woodward TE, Smadel IE, Ley HL, Green R. Preliminary report on beneficial effectof chloromycetin in treatment of typhoid fever. Ann Intern Med 1948;29:131-4.

3 Colquhoun J, Weetch RS. Resistance to chloramphenicol during treatment oftyphoid fever. Lancet 1950;2:621.

4 Vazquez V, Calderon E, Rodriquez RS. Chloramphenicol resistant strain ofSalmonella typhosa. N Engl } Med 1972;286:1220.Datta N, Richards H, Datta C. Salmonella typhi in vivo acquires resistance to bothchloramphenicol and co-trimoxazole. Lancet 1981;1:1181-3.

6 Samantray SK. Typhoid fever resistant to furazolodine, ampicillin, chloramphenicoland co-trimoxazole. Indian} Med Sci 1979;33:1-3.

7 Ling J, Chau PY. Plasmids mediating resistance to chloramphenicol, trimethoprimand ampicillin in Salmonella typhi strains isolated from Southeast Asian region.} Infect Dis 1984;149:652-3.

8 Anderson ES, Smith HR. Chloramphenicol resistance in the typhoid bacillus.8M} 1972;3:329-31.

9 Anderson ES. The problem and implications of chloramphenicol resistance in thetyphoid bacillus. J Hyg 1975;74:289-99.

10 Edelman R, Levine MM. Summary of an international workshop on typhoid fever.Rev Infect Dis 1986;8:329-49.

11 Schroeder SA. Interpretation of serological tests for typhoid fever. }AMA 1968;206:839-41.

12 Lewis, Chatopadhyaya et al. Serological tests in typhoid fever. Are they of use?Infect Dis 1988;2:391.

13 Senewiratne B, Senewiratne K. Reassessment of the Widal test in the diagnosis oftyphoid. Gastroenterology 1977;73:233-6.

14 Pang T, Puthucheary SO. Significance and value of the Widal test in the diagnosisof typhoid fever in an endemic area. J Clin PathoI1983;36:471-5.

15 Agarwal KC, Panhotra BR, Mahanta J, Arya VK, Garg RK. Typhoid fever due tochloramphenicol resistant Salmonella typhi associated with R plasmid. Indian}Med Res 1981;73:484-8.

16 Shridhar H, Macaden R, Devi MC, Bhat P. Chloramphenicol resistant S. typhi inBangalore. Indian} Med Res 1983;78:314-18.

17 Sarkar AK, Ganguly S, Ganguly S. Recent outbreak of chloramphenicol resistanttyphoid fever in West Bengal. J Indian Med Assoc 1991;89:257-9.

18 Prabakar H, Kaur H, Lal M. Prevalence of multidrug resistant S. typhi in Ludhiana,Punjab. Indian J Med Sci 1996;50:277-9.

19 Jesudason MV, Malathy B, John TJ. Trend of increasing levels of minimum inhibi-tory concentration of ciprofloxacin to Salmonella typhi. Indian J Med Res 1996;103:247-9.