Typhoid fever is caused by Salmonella enterica serovar Typhi (hereinafter referred to as “Typhi”) and is characterized by prolonged fever, headache, nausea, loss of appetite, constipation, or diarrhea [1]. Globally, South Asia accounts for more typhoid fever cases each year than any other geographic region [2,3]. An estimated 11-21 million cases of typhoid fever and approximately 128,000-161,000 deaths are reported annually [3]. Among countries in South Asia, Pakistan has the highest estimated incidence rate of typhoid fever (493.5 per 100,000 persons/year) [4,5]. Treatment of enteric fever has been complicated by the development and rapid global spread of typhoidal organisms resistant to ampicillin, trimethoprim-sulfamethoxazole, and chloramphenicol. Additionally, the development of increasing resistance to fluoroquinolones is a growing challenge [6]. Since the extensively drug resistant (XDR) Typhi outbreak began in Hyderabad in late 2016 [7], typhoid fever cases in Pakistan increased, particularly in Hyderabad and other parts of Sindh province; this is attributed to the circulating XDR Typhi strain [8].

The most common serious and fatal complication of typhoid fever is intestinal perforation [9]. Even though many advancements have been made in patient care, morbidity and mortality rates remain high among patients with typhoid perforation, especially in low- and middle-income countries [10,11]. The high incidence of typhoid ileal perforations in developing countries is reportedly associated with delayed diagnosis and treatment and the emergence of antibiotic resistance. Once complications appear, patients require hospitalization and sometimes tertiary care [12]. Postoperative complications and mortality are common [13].

In November 2016, a multidrug-resistant Typhi strain that is resistant to ampicillin, chloramphenicol, and trimethoprim-sulfamethoxazole and additionally to ciprofloxacin and ceftriaxone was identified in patients hospitalized in Hyderabad, Pakistan. This XDR Typhi strain belongs to the H58 lineage and includes a plasmid-mediated blaCTX-M-15 extended-spectrum β-lactamase [14]. Typhoid cases have since been identified in Hyderabad, Karachi, and in other districts of Sindh province, and over 5372 cases were reported by December 2018. As of that month, no isolates resistant to azithromycin or a carbapenem had been reported. Despite the large number of cases, little information on mortality and morbidity attributable to XDR Typhi infections has been available. To address this knowledge gap, this study aimed to assess the morbidity and mortality associated with XDR and non-XDR Typhi infections in Hyderabad, Pakistan.

The medical records of all hospitalized patients with culture-confirmed typhoid fever between October 1, 2016, and September 30, 2018, at the 5 major hospitals in Hyderabad, including Liaquat University City, Jamshoro, Shah Bhittai, Qasimabad, and Peratabad Hospitals, were reviewed. Subsequently, the pediatric, medical, and surgical ward records of all identified cases were also reviewed. We obtained ethics approval from the institutional review board of Liaqat University of Medical Sciences, Jamshoro, and administrative approvals from all the participating hospitals.

We collected data on demographic characteristics, date of fever onset, illness duration and antibiotic history before hospitalization, physical examination findings, results of serological and microbiologic tests, treatments prescribed or administered in hospital, date of recovery, duration of hospitalization, complications (including ileal perforation, septicemia, cholecystitis, pneumonia, jaundice, encephalopathy, meningitis, and deaths), and antimicrobial susceptibility of Typhi isolates. XDR Typhi cases were defined as those whose isolates were resistant to the 5 aforementioned agents. Ileal perforations were diagnosed if biopsy reports indicated “a hollow viscous perforation of the terminal part of the ileal wall” and pus culture sensitivity reports indicated an XDR status among patients with ileal perforation.

Data were recorded using a specially designed questionnaire and transferred to a computerized database for statistical analysis in Epi Info (Centers for Disease Control and Prevention). The data are expressed as mean, median, and percentage values. The chi-square test was used to compare percentages and the Mann–Whitney U test was used to compare medians. A P value of <.05 was considered significant.

A total of 1452 cases of culture-confirmed typhoid fever were identified from in-patient departments of 5 hospitals in Hyderabad during the study period. Patient ages ranged 1-80 (mean 14, median 22) years. More than half (n=791, 55%) of the patients were aged between 2 and 10 years. Almost two-thirds (n=915, 63%) of patients were male. The total number of monthly cases ranged from 2 in October 2016 to 98 in May 2018. Typhi isolates from approximately two-thirds of cases (n=947, 66%) were XDR, while the remainder (n=505, 35%) were non-XDR. Patients with non-XDR typhoid belonged to a slightly younger age group. Males accounted for similar proportions of XDR and non-XDR cases (64% and 62%, respectively) (Table 1).

The signs and symptoms in patients with XDR typhoid and those with non-XDR typhoid on hospital presentation were very similar (Table 2).

The duration of illness between fever onset and hospitalization ranged 5-28 (median 20) days for patients with XDR typhoid, and 5-30 (median 13) days for those with non-XDR typhoid (P<.001). The conventional antibiotics taken before hospitalization among patients with XDR typhoid were ceftriaxone and ampicillin, whereas those among patients with non-XDR typhoid were cefixime, cephalosporin, and quinolones. In the hospitals, all patients with XDR typhoid were treated with a carbapenem or macrolides with a median hospitalization duration of 9.5 (range 2-25) days, while those with non-XDR typhoid were treated with cephalosporin and cefixime (n=483, 33%) and with ceftriaxone and ampicillin (n=22, 1.5%) with the median hospitalization duration of 6.5 (range 2-15) days.

Overall, ≥1 complications were reported in 360 (38%) patients with XDR and in 89 (18%) patients with non-XDR typhoid (P<.001). Ileal perforation was the most commonly reported complication in both patients with XDR (n=210, 23%) and those with non-XDR (n=71, 14%) typhoid (P<.001). Septicemia was the second-most commonly reported complication among both patients with XDR (n=97, 11%) and those with non-XDR (n=9, 1.7%) typhoid (Table 3).

The median duration of prehospitalization illness was longer among patients with complications (range 7-30 days, median 22 days) than among those without complications (range 5-25 days, median 6 days) (P<.001). Overall, mortality was documented in 17 of 947 (1.8%) patients with XDR Typhi infections and 3 of 505 (0.6%) patients with non-XDR Typhi infections (P=.06).