Six thousand five hundred and twenty seven stool samples from patients with diarrhea presenting in Lok Nayak Hospital, New Delhi were received over a period of three years (2012-14) in the Microbiology Laboratory, Maulana Azad Medical College, New Delhi, India. An episode of diarrhea was defined as three or more loose or watery stools per day. Stool samples were processed for parasitological and bacteriological identification and antimicrobial susceptibility of bacterial isolates.

Stool samples were examined grossly for color, consistency, presence or absence of blood, mucus and worms. Routine stool microscopic examination of saline and iodine preparation was done for red blood cells, pus cells, trophozoites and cysts of protozoa and ova of helminthes. Modified acid fast stain [4] was done to visualize oocysts of Cryptosporidium and Isospora belli.

Stool samples were cultured on several media for maximal yield. Sample was inoculated directly and after enrichment in selenite F broth and APW (Alkaline peptone water) onto MacConkey’s agar, Xylose Lysine Desoxycholate agar, Deoxycholate Citrate Agar and Bile salt agar. The plates were incubated for 18-24 hours at 37ºc. Suspected colonies were subjected to preliminary tests like Gram’s staining, hanging drop for motility, catalase and oxidase tests. Their identity were established by a battery of biochemical tests like: fermentation of a variety of sugars, indole test, citrate utilization, urease production, MR and VP tests, production of H₂S on TSI and lysine and ornithine decarboxylase tests and arginine dihydrolase test.

All the organisms identified as Shigella spp., Vibrio cholerae or Salmonella on the basis of colony characteristics and biochemical reactions were confirmed by the slide agglutination test with specific antisera (Denka Seiken Co., Ltd., Tokyo, Japan). E. coli isolates from children below 5 years of age were also serotyped to identify diarrheagenic E. coli (Denka Seiken Co., Ltd., Tokyo, Japan).

The antimicrobial susceptibility testing was performed for all bacterial isolates by using the modified Stokes’ disc diffusion method [5] for the following six antimicrobial agents- Nalidixic acid (30μg), Amoxicillin (10μg), Amikacin (30 μg), Gentamicin (10 μg), Ciprofloxacin (5 μg), Cefotaxime (75 μg). These discs were either prepared in laboratory or commercially procured from HiMedia, Mumbai, India.

Out of 6527 enrolled patients, 3823 (58.57%) were under 5 y of age, 1505 (23.05%) were between 5 y and 12 y and 1199 (18.37%) were more than 12 y of age. On gross examination, 1209 (18.52%) samples were formed in consistency, 3215 (49.25%) were semi-formed and 2103 (32.22%) were liquid. Presence of mucus in stools was observed in 931 (14.26%) samples, while approximately 4.92% (321/6527) samples contained grossly visible blood.

Bacterial pathogens were isolated from 581 (8.90%) of the total of 6527 stool samples. These included the diarrheagenic E. coli,that were identified by serotyping only in children who were below five years of age. Out of these 3823 samples, 194 (5.07%) yielded enteropathogenic Escherichia coli (EPEC). Enterotoxigenic Escherichia coli (ETEC), Entero aggregative E. coli (EAEC) and enteroinvasive Escherichia coli (EIEC) were isolated in considerably lesser numbers [Table/Fig-1]. The most common EPEC, ETEC, EAEC and EIEC serotypes isolated were O128, O147, O44 and O164, respectively.

Vibrio cholerae were found in 159 out of 6527 (2.44%) samples. All Vibrio cholerae were serotyped as O1 Ogawa. Other strains of Vibrio cholerae and Vibrio parahemolyticus were not isolated in any sample. A total of 126 (1.93%) Shigella spp. was isolated in cases with dysentery. Shigella flexneri was the most common isolated species and it was present in 0.96% cases, followed by Shigella boydii, Shigella dysenteriae and Shigella sonnei. Salmonella Typhi was isolated in 7 cases and Salmonella Typhimurium in 6 cases. Aeromonas hydrophila was found in 3 cases [Table/Fig-1].

Parasites were identified in 312 (4.78%) patients as the aetiological agent of diarrhea. Giardia intestinalis, Ascaris lumbricoides and Entamoeba histolytica comprised the major parasitic pathogens (2.27%, 1.15% and 0.64% respectively) [Table/Fig-1]. Cryptosporidium and Isospora were identified in 10 (0.15%) and 9 (0.13%) cases respectively and all of these patients were retrospectively found to be HIV positive.

Susceptibility levels of bacterial pathogens to various antimicrobial agents are shown in [Table/Fig-2]. All bacterial isolates were highly sensitive to amikacin and poorly sensitive to nalidixic acid. E. coli and Vibrio cholerae showed 89.64% and 96.22% sensitivity to amikacin respectively. Some pathogens like Shigella boydii and Shigella dysenteriae were completely resistant to nalidixic acid. Shigella dysenteriae was completely resistant to amoxicillin. Activity of ciprofloxacin was seen to fall against E. coli and Shigella spp. High resistance to third generation cephalosporins was seen among diarrheagenic E. coli and Shigella spp.

Our hospital is a centrally located, 1676 bedded tertiary care hospital with thickly populated catchment areas. Most of the patients fall in the low to middle socio-economic groups. This explains the large sample load that we received over a period of two years.

Diarrhea is a major cause of morbidity and mortality in India and other developing countries of the tropical zone [9–12]. Thus, there is a constant need to keep the knowledge updated with regards to pathogens associated with diarrheal illness and their antimicrobial sensitivity pattern for providing effective health care facilities. The empiric antimicrobial treatment, if required should be directed against locally prevalent pathogens.

One of the objectives of the study was to identify the recent papers (after year 2000) reporting aetiology of infectious diarrhea in India involving the general population as a whole and compare them with present findings. Thus, we searched the PubMed/Medline, Google scholar and various popular journals to look for various combination of following terms: ‘diarrhea’, ‘gastroenteritis’, ‘enteropathogens’, ‘aetiology’, ‘epidemiology’, ‘India’, etc. Studies enrolling only patients with clinical signs of dysentery, i.e. blood in the stool, studies conducted in special populations such as infants, children, travellers, patients hospitalized for other specific reasons, or only HIV positive persons were excluded.

Majority of the studies conducted in India, since the year 2000, have restricted themselves to narrow groups (such as infants, children, HIV positive patients, etc.). Despite thorough search, we were able to find very few studies meeting the exclusion and inclusion criteria [Table/Fig-1]. Paucity of similar studies has also been reported by Walker CLF et al., [11] on a global scale. They were able to identify only 22 studies worldwide, meeting exclusion and inclusion criteria which were similar to ours.

The data from various Indian studies was then compiled to make an attempt to identify the most common bacterial or parasitic agent responsible for diarrhea in India. The compiled sample size for all the national studies was 13164 [Table/Fig-1]. Vibrio cholerae was found to be the most common bacterial agent in Indian studies, responsible for 9.99% cases. This was followed by diarrheagenic E. coli (8.27%) and Shigella spp (2.51%) as the second and third most frequent bacterial agents. The findings of the present study differed slightly; showing that in our region diarrheagenic E. coli (7.31%) is most common, followed by Vibrio cholerae (2.44%) and Shigella spp (1.91%).

In this study, serotyping for diarrheagenic E. coli was performed only for children under 5 years of age, keeping in view the fact that E. coli causes infection mostly in children and also due to limited availability of antisera. Out of 3823 under-5 year cases, 280 (7.31%) cases were infected with diarrheagenic Escherichia coli. This finding is in accordance with several studies from other developing countries as well [6,10,13–15]. A systematic review concluded that rotavirus, calicivirus, enteropathogenic and enterotoxigenic E. coli cause more than half of all diarrheal deaths in children <5 y in the world, thus emphasizing the role of E. coli in childhood diarrhea [16]. There were no major differences between the susceptibility profiles of EPEC, ETEC, EAEC and EIEC [Table/Fig-2].

Vibrio cholerae is endemic in India and is characterized by yearly and seasonal outbreaks. In our study, V. cholerae was found to be the second most common cause of diarrhea, with almost three-fourth of strains being isolated during the months of July to September. All of the isolated Vibrio cholerae strains were of the Serogroup O1, serotype Ogawa. However, Das et al., found 56 of 96 strains isolated in 2004 were of serotype Inaba [6]. These strains were largely sensitive to ciprofloxacin, amikacin, gentamicin and cefotaxime and resistant to nalidixic acid and amoxicillin [Table/Fig-2], in agreement with previous studies from Delhi [6] Chandigarh [17] and Calcutta [18].

In industrialized countries, shigellosis occurs mostly due to Shigella sonnei while in the developing countries, most cases are due to Shigella flexneri, Shigella dysenteriae and Shigella boydii [19]. In the present study, Shigella flexneri was the most common causative agent followed by Shigella boydii,Shigella dysenteriae and Shigella sonnei. Poor sensitivity to nalidixic acid and amoxicillin and rising resistance to ciprofloxacin and third generation cephalosporins raises concerns and demands more restricted use of these antibiotics to control further increase in resistance [Table/Fig-2].

Compiled data from recent national studies [Table/Fig-1] suggested that among the protozoans, Giardia intestinalis (3.57%), Cryptosporidium spp. (1.34%) and Entamoeba histolytica (1.15%) were most frequent offenders. Helminthic infections such as Ascaris lumbricoides (0.76%), Hookworm (0.18%) and Trichuris trichura (0.08%) were less common findings. Findings of the study at our centre were slightly different with respect to the incidence of various parasitic infections.

Cryptosporidium and Isospora are infrequent finding in India accounting for 1.34% and 0.07% cases of diarrhea [Table/Fig-1]. Another study found that 70 (2.7%) of 2,579 (2.7%) children were found be positive by microscopic methods, and Cryptosporidium hominis was the most frequent species as determined by genotypic methods [20]. In the current study, these were identified in 0.15% and 0.13% cases, respectively. All these patients were HIV positive, thus suggesting that HIV positive individuals should be examined for these parasites.

This study was limited by the fact that there is very little recent published data from India documenting the exact prevalence of bacterial and parasitic diarrhea causing agents that met our inclusion and exclusion criteria, encompassing the general population and not restricting to specific groups. Moreover, use of different methods and use of special media for specific organisms such as Campylobacter spp. may have resulted in non-uniform reporting of results. Nevertheless, inclusion of large compiled sample size of 13164 specimens compensates for the limited number of studies available to a certain extent.

Analysis of recent nationwide studies revealed V. cholerae was the most common bacterial/parasitic agent of diarrhea across all populations, being followed by diarrheagenic E. coli and Giardia intestinalis.

This study also documents the antimicrobial susceptibility of bacterial agents to different antibiotic agents. Amikacin and third generation cephalosporins retain their activity as an anti-diarrheal agent. However, their indiscriminate use may cause development of resistance to these therapeutic agents in the near future. Antimicrobials should be resorted to only in severe cases, who do not improve despite rehydration therapy. Proper maintenance of sanitation and hygiene is also useful to check transmission of enteric pathogens.