Original Article#

Enteric Fever in Mumbai – Clinical Profile, Sensitivity

Patterns and Response to Antimicrobials
S Jog*, R Soman**, T Singhal***, C Rodrigues+, A Mehta+, FD Dastur**

Abstract
Aims : Enteric fever is endemic in Mumbai and its diagnosis poses several problems. Our main aim was
to study the clinical profile, haematological features of culture proven typhoid cases, the antimicrobial
susceptibility pattern of the isolates and the time to defervescence with the treatment received.
Material and Methods : This was a retospective chart review of all cases of culture proven enteric fever carried
out at a tertiary care private hospital in Mumbai over the period January 2003 to September 2005.
Results : Culture positivity in our study was 52.6%. Sixty one percent of the isolates were Salmonella typhi
while 39% were Salmonella paratyphi A. An absolute eosinopenia was seen in 76.9% of the patients. Before
being admitted to the hospital, 46.2% received antibiotics. The mean time to defervescence in patients who
received prior antibiotics was 4.5 days while that in those who did not receive prior antibiotics was 5.1
days.
Conclusions : A high culture positivity despite prior or ongoing antibiotic treatment was seen. Absolute
eosinophil count of 0% could be an important marker of typhoid. High prevalence of nalidixic acid resistance,
a marker of resistance to fluoroquinolones was observed. Combination treatment was not found to be superior
to treatment with a single antibiotic. ©

INTRODUCTION Fluoroquinolones when first introduced in early 1990’s

were very effective but the past decade has seen a
E nteric fever is a systemic illness caused by Salmonella
enterica serotype typhi or paratyphi A/B. In India the
disease is endemic with an incidence ranging from 102
progressive increase in the MICs of ciprofloxacin and
high incidence of clinical failure to quinolones. The
to 2219 per 100,000 population.1 It results in considerable beta lactams such as cefixime and ceftriaxone are now
morbidity, absenteeism and resource utilization. being increasingly used but these are expensive drugs
and are associated with a long time to defervescence
Diagnosis of enteric fever is fraught with problems.
and high rates of relapse. There have also been sporadic
History, physical findings and fever pattern are
reports of high-level resistance to ceftriaxone in S. typhi
suggestive but can neither confirm nor exclude typhoid.
and S. paratyphi.2 Experience with new drugs such as
Blood culture is the ‘gold standard’ for diagnosis and
azithromycin is at present scanty.5-7
also gives information about antibiotic sensitivity of the
isolate; however the cost of cultures, lack of “culture To sum up, enteric fever raises several issues of
of cultures” and administration of prior antibiotics diagnosis and treatment. There is a scarcity of studies in
are impediments in this diagnostic approach. The literature addressing these issues. Hence a retrospective
Widal test is very commonly used in Indian set up review of patients of enteric fever at our hospital was
but has very variable sensitivity and specificity and carried out to solve some of these pertinent clinical
problems in interpretation. Therapy of enteric fever problems.
is becoming more complicated and expensive with Aims and Objectives
time. By the end of 1990s, Salmonella enterica developed
The primary aim of the study was to study the
resistance simultaneously to all first line drugs like
sensitivity patterns of S. enteritica and response to
chloramphenicol, cotrimoxazole and ampicillin.
antimicrobial therapy. The secondary aims were to
study the clinical and laboratory profile of patients with
*Fellow, Dept. of Infectious Diseases; **Consultant Physician;
***Consultant Paediatrician; +Consultant Microbiologist,
culture proven enteric fever.
P.D.Hinduja National Hospital and Medical Research Centre, Veer
Savarkar Marg, Mahim, Mumbai 400016, India. MATERIAL AND METHODS
Received : 2.1.2007; Revised : 18.2.2008; Accepted : 3.3.2008
#Rapid Publication
This was a retrospective chart review of all cases
of enteric fever carried out at a tertiary care private
© JAPI  •  VOL. 56  •  APRIL 2008 www.japi.org 237
hospital in Mumbai, India. The records of all patients 1 of these 7 was culture proven, 2 were culture negative
discharged from our hospital between January 2003 and no details were available for the rest.
and September 2005 with a diagnosis of enteric fever, The mean white blood cell (WBC) count was 6358/
typhoid fever or paratyphoid fever were assessed for cumm with a range from 1050/cumm to 15,120/
suitability for inclusion in our study. These records cumm. Eighty five percent had the WBC count within
were retrieved from the Medical Records Section of the the normal range (4000 –11000/cumm) while 11.4%
hospital after going through the computer records using patients had leucopenia (WBC count < 4000/cumm).
‘enteric fever, typhoid fever or paratyphoid fever’ as Leucocytosis (WBC count > 11000/cumm) was seen in
discharge diagnosis in the search criteria. Only culture only 4 patients. Absolute eosinopenia (0% eosinophils)
proven cases of enteric fever were included in the study. was seen in 76.9% patients. The mean platelet count of
Others were considered as clinically diagnosed typhoid the study patients was 2,04,800/ cumm (range 31,000
and were excluded from the study. They were used to to 4,88,000). Thrombocytopenia (platelet count < 1.5
calculate the culture positivity. Clinical, laboratory and lacs/ cumm) was seen in 25.9%. Hyperbilirubinemia (>
treatment information was extracted from the medical 1 mg/dl) was seen in 28.7% while the ALT was elevated
records on a detailed proforma and analyzed. The two- (> 60 IU/ml) in 43% patients.
sample t test was used to compare continuous variables
Seventy-three of the 119 study patients (61%) were
and the chi square test was used to compare categorical
positive for Salmonella typhi while the rest had Salmonella
variables. A p value of less than 0.05 was considered
paratyphi A. The 73 isolates of Salmonella typhi included
significant.
70 from blood culture, 2 from bone marrow and 1 from
The Ethics Committee of the hospital had no objection stool culture. All the Salmonella paratyphi A were isolated
for retrospective data analysis, as the data collected from blood cultures. The cultures were sent after a mean
was based on routine clinical practice. Further, the period of 11 days after the onset of fever. Table 1 shows
committee had no objection to publish a paper based the sensitivity pattern of the isolates year wise over the
on this retrospective data, provided the patient identify study period. There is no significant difference between
was not revealed. the year wise antimicrobial susceptibility patterns of
that of S typhi and S paratyphi A. Data for azithromycin
RESULTS is not included as sensitivity to azithromycin was not
A total of 226 patients received a discharge diagnosis tested during the study period. None of the isolates
of enteric fever/ typhoid fever/ paratyphoid fever of Salmonella paratyphi A showed block resistance to
during the period January 2003 to September 2005. Out ampicillin, chloramphenicol and cotrimoxazole.
of these 226, 119 (52.6%) were culture proven cases of Widal test results were available for 64 of 119 patients.
enteric fever and were included in the analyses. Widal test was positive (defined as S. typhi O antigen
Seventy-four of the 119 study patients were male >120 and either S. typhi H or S. paratyphi H antigen titres
and 45 were female. The age ranged from 7 months to >120) in 24 of 64 patients , (48.4%). The mean duration
66 years, the mean age being 21.7 years. of fever at which Widal test was positive was 10.7 days
All patients had fever prior to admission; the median (95% CI 8.3, 13.1 days).
duration of fever prior to admission was 7 days, the Fifty-five patients (46.2%) received antibiotics before
range being 2 to 90 days. High-grade fever was seen being admitted to the hospital. Beta lactams (cefuroxime,
in 66.3% and chills were present in 57.9%. Vomiting, cefadroxyl, cefixime, ceftriaxone, amoxycillin,
abdominal pain and loose stools were the most common cefotaxime, amoxycillin-clavulanate) were taken by 20
associated symptoms seen in 42%, 33.6% and 31% patients before admission while quinolones were used in
patients respectively. Only two patients complained 14 patients. Three patients gave history of having taken
of constipation. None of the patients had relative azithromycin. Twelve patients had taken a combination
bradycardia. Hepatosplenomegaly was seen in 12.6% of antibiotics including beta lactams, quinolones,
patients. Only hepatomegaly was seen in 15.9% while aminoglycosides, azithromycin and chloramphenicol.
only splenomegaly was seen in 7.5% patients. Seven Data on the duration of prior antibiotic use was available
patients gave history of enteric fever in the past. Only only in 20 of the 55 cases and the mean duration of prior
Table 1 : Year wise sensitivity pattern of the Salmonella enteritica isolates
        2003              2004               2005 Overall
S. typhi S. par A S. typhi S. par A S. typhi S. par A S eneteritica
29n (%) 15n (%) 24n (%) 15n (%) 20n (%) 16n (%) 119n (%)
S to A, CH, TS 28 (96) 15(100) 23 (96) 15 (100) 17 (85) 16 (100) 114 (96)
S to NA 9 (31) 2 (13) 7 (30) 2 (13) 5 (25) 0 (0) 25 (21)
S to ceftriaxone 100%
S = sensitive, A = ampicillin, CH = chloramphenicol, TS = TMP-SMX, NA = nalidixic acid

238 www.japi.org © JAPI  •  VOL. 56  •  APRIL 2008

antibiotic use was 3.9 days (range 1 to 10 days). megakaryocytes and increased phagocytic activity of
Ceftriaxone was the most common antibiotic used histiocytes in the bone marrow.11
to treat patients in hospital; 74 of 119 patients (62.1%). As many as 46.2% patients in our study received
A combination of ceftriaxone (2 gm bd IV) and antibiotics either single or in combination for as long as
azithromycin (500 mg od PO) was used in 16 patients 10 days before being admitted to the hospital. Despite
(13.4%). Rest of the patients (25%) received various this they were still culture positive. The high yield from
other antibiotics singly or in combination. The mean blood cultures despite antibiotics could have been as a
duration of receipt of antimicrobials after hospitalization result of processing of the blood cultures by the BACTEC
was 11 days. method, which uses soybean casein digest broth, yeast
The mean time to defervescence defined as time extract, animal tissue digest, sucrose, hemin, menadione,
period in days from the day of onset of the antibiotic pyridoxal hydrochloride, sodium bicarbonate and
treatment in the hospital to the disappearance of fever sodium polyanethol sulphonate as a culture medium
was calculated for various patient groups. The mean optimised with cationic exchange resin. The usual
time to defervescence in the group of patients who had perception is that culture positivity falls dramatically
received antibiotics prior to admission was 4.5 days with prior use of antibiotics9,10,12 and often dissuades
(95% CI 3.9, 5.1 days) while in the patients who did clinicians from sending blood cultures in patients with
not receive prior antibiotics was 5 days (95% CI 4.3, 5.7 pyrexia. However results of our study advocate that
days) , p 0.2 (NS). blood cultures should be sent in suspected enteric fever
even if the patient is on antimicrobials.
The overall mean time to defervescence when
ceftriaxone alone was used as therapy was 4.2 days. It Interestingly 40% of the isolates in our study were
was 4.2 days in 2003, 4.4 days in 2004 and 4.2 days in S paratyphi A as against 20% in literature.13 This may
the year 2005 (p = 0.84). The mean time to defervescence be a consequence of increasing immunization with
in the patients who received ceftriaxone in hospital did the Vi antigen vaccine that does not protect against S.
not differ significantly between those that had received paratyphi.14 The sensitivity pattern of the isolates obtained
prior antibiotics and those who did not receive prior in this study is in accordance with the other studies
antibiotics (4.4 days versus 5 days respectively). from India.1,15 There is high prevalence of nalidixic acid
resistance and return of sensitivity to chloramphenicol,
The mean time to defervescence in those patients who ampicillin and cotrimoxazole. We did not observe any
received a combination of ceftriaxone and azithromycin resistance to third generation cephalosporins as in other
was 5.1 days and did not differ significantly from those studies by Chowta et al,1 Safdar et al,16 etc. Chande et al
who received ceftriaxone alone (p 0.06). observed resistance to cefotaxime in one isolate in their
study from central India.17
DISCUSSION
The mean fever clearance time with ceftriaxone used
This study is one of the largest retrospective studies
as single therapy observed in our study was 4.2 days
on enteric fever. Other large series include those by
(95% CI 3.7, 4.7 days) as against 6.1 days in literature.4
Chowta et al1 and Walia et al.8
No change in the time to defervescence with ceftriaxone
Most of the clinical symptoms and signs reported by over the years was seen in our study (4.2 days in the year
us are similar to those reported earlier.1,4,9 Very prolonged 2003, 4.4 days in 2004 and 4.2 days in 2005). Surprisingly,
fever lasting more than 90 days, seen in 2 of our patients receipt of antibiotic therapy prior to admission was
was unusual for enteric. This may have been due to an found to have no significant impact on the time to
additional cause operating in the earlier part of illness. defervescence.
Relative bradycardia and constipation considered
There has been no retrospective/prospective study
to be salient features of enteric were not seen/very
evaluating the efficacy of combination therapy for
infrequently seen in our study. Few other studies have
enteric fever. Results of our study however suggest
also found these to be inconsistent features of enteric
that combination therapy may not be superior to single
fever.1 Four patients in our study had leucocytosis, a
drug therapy, as we did not observe any significant
laboratory finding that is believed to cast a doubt on its
difference in the time to defervescence in those patients
being a differential diagnosis for pyrexia. Conversely,
who received ceftriaxone alone or in combination with
absolute eosinopenia, (seen in 77% of our patients) can
azithromycin.
be used as a pointer of enteric when a complete blood
count is done in a patient with fever. Deshmukh et al in Limitations
their study on paediatric patients with bacteriologically The main limitation of this study is its retrospective
and/or serologically diagnosed typhoid fever found nature and that it was not protocol driven. Patients
absolute eosinopenia in 71.4% of patients.10 Leucopenia, were receiving various antimicrobials singly or in
eosinopenia, thrombocytopenia and anemia in combination for varying periods prior to hospitalization,
enteric can be attributed to the myeloid maturation which could impact the time to defervescence. Also the
arrest, decrease in the number of erythroblasts and inability of the study to detect a difference in efficacy
© JAPI  •  VOL. 56  •  APRIL 2008 www.japi.org 239
of various antimicrobial regimes chiefly single versus for the treatment of uncomplicated typhoid fever in children
combination therapy may be due to the small sample and adolescents. Clin Infect Dis 2004;38:951-857.
size. 6. Clinh NT, Parry CM, Ly NT, Ha HD, Thong MX, Diep TS, Wain
J, White NJ, Farrar JJ. A randomized control comparison of
The efficacy of various antimicrobials in enteric fever azithromycin and ofloxacin for treatment of multidrug – resistant
can be best studied by a randomized controlled trial in or nalidixic acid – resistant enteric fever. Antimicrob Agents
a large number of antimicrobial naïve culture positive Chemother 2000;44:1855-59.
patients. However it is virtually impossible to recruit 7. Butler T Sridhar CB, Daga MK, Pathak K, Pandit RB, Khakhria K,
antimicrobial naïve culture positive patients. Potkar CN, Zelasky MT, Johnson RB. Treatment of typhoid fever
with azithromycin versus chloramphenicol in a randomized
With all the limitations notwithstanding this study multicentre trial in India. J Antimicrob Chemother 1999;44:243-
is the largest study of culture proven typhoid. Some 50.
important conclusions which can be drawn from the 8. Walia M, Gaind R, Mehat R, Paul P, Aggarwal P, Kalaivani M.
study include importance of absolute eosinopenia as Current perspectives of enteric fever : A hospital – based study
a diagnostic marker of typhoid, high culture positivity from India. Ann Trop Paediatr 2005;25:161-74.
despite receipt of prior antibiotics, high prevalence of 9. Kadhiravan T, Wig N, Kapil A, Kabra SK, Renuka K, Misra A.
Clinical outcomes in typhoid fever : Adverse impact of infection
nalidixic acid resistance (79%), return of susceptibility to
with nalidixic acid – resistant Salmonella typhi. BMC Infectious
chloramphenicol (96%), 100% sensitivity to ceftriaxone Diseases 2005;5:37.
and non superiority of combination therapy versus 10. Deshmukh CT, Nadkarni UB, Karande SC. An analysis of
single agent therapy. Urgently needed are well-designed children with typhoid fever admitted in 1991. J Postgrad Med
randomized controlled trials to compare the efficacy of 1994;40:204-7.
various antibiotics in inpatient and outpatient therapy 11. Khosla SN, Anand A, Singh U, Khosla A. Haematological profile
of enteric fever. in typhoid fever. Trop Doct 1995;25:156-8.
12. Ananthanarayan R, Paniker CKJ. Textbook of Microbiology. 6th ed.
REFERENCES (Orient Longman Private Ltd., Hyderabad, India) 2000:267-80.
1. Chowta MN, Chowta NK. Study of Clinical Profile and Antibiotic 13. Crump JA, Luby SP, Mintz ED. The global burden of typhoid
Response in Typhoid Fever. Indian J Med Microbiol 2005;23:125- fever. Bull World Health Organ 2004;82:346-53.
27.
14. Rodrigues C, Shenai S, Mehta A. Enteric fever in Mumbai : the
2. House D, Wain J, Ho VA, Diep TS, Chinh NT, Bay PV, Vinh good news and the bad news. Clin Infect Dis 2003;36:535.
H, Duc M. Parry CM, Dougan G, White NJ, Hien TT, Farrar
JJ. Serology of typhoid fever in an area of endemicity and its 15. Gautam V, Gupta N, Chaudhary U, Arora DR. Sensitivity pattern
relevance to diagnosis. J Clin Microbiol 2001;39:1002-7. of Salmonella serotypes in Northern India. Brazilian J Infect Dis
2002;6:281-87.
3. Olsen SJ, Pruckler J, Bibb W, Nguyen TM, Tran MT, Sivapalasingam
S, Gupta A, Phan TP, Nguyen TC, Nguyen VC, Phung DC, Mintz 16. Safdar A, Kaur H, Elting L, Rolston KV. Antimicrobial
ED. Evaluation of rapid diagnostic tests for typhoid fever. J Clin susceptibility of 128 14. Salmonella enterica serovar typhi
Microbiol 2004;42:1885-9. and paratyphi A isolates from northern India. Chemotherapy
2004;50:88-91.
4. Parry CM, Hien TT, Dougan G, White NJ, Farrar JJ. Typhoid
Fever. New Engl J Med 2002;347:1770-82. 17. Chande C, Shrikhande S, Kapale S, Agrawal S, Fule RP. Change
5. Frenek RW Jr, Mansour A, Nakhla I, Sultan Y, Putnam S, in antimicrobial resistance pattern of Salmonella typhi in central
Wierzba T, Morsy M, Knirsch C. Short – course azithromycin India. Indian J Med Res 2002;115:248-50.

Announcement
Eligibility Criteria for the Award of Fellowship of
Indian College of Physicians
Full Format available on API and JAPI Website - www.apiindia.org and www.japi.org

Dr. Sandhya Kamath Dr. Falguni Parikh

Hon General Secretary, Jt Secretary,
API & ICP API & ICP

240 www.japi.org © JAPI  •  VOL. 56  •  APRIL 2008