7 pages

Acute infectious diarrhoea in adults: identifying clinical parameters associated with specific pathogens

Please download to get full document.

View again

of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Acute infectious diarrhoea in adults: identifying clinical parameters associated with specific pathogens
  Hong Kong Journal of Emergency Medicine Acute infectious diarrhoea in adults: identifying clinical parametersassociated with specific pathogens SSW Chan, KC Ng, PKW Lam, DJ Lyon, WL Cheung, TH Rainer Introduction:   Infectious diarrhoea may be caused by viral, bacterial or protozoan agents. The objective of this study was to explore the possibility of correlating presenting clinical and demographic features withthe specific types of stool pathogens subsequently identified. Materials & Methods:   A retrospective studywas performed in the setting of an Accident & Emergency (A&E) department of an urban acute-carehospital in Hong Kong. The inclusion criteria were adults (age ≥ 16); presentation with features of acuteinfectious diarrhoea; treated as out-patients with or without observation; and with stool cultures requestedfrom A&E. All consecutive culture-positive cases (n=130) satisfying the above criteria were included. Thecontrol-group (n=119) consisted of a random sample of culture-negative cases during the same study period.For each of the six pathogens identified, statistical analyses were performed to compare differences inclinical features amongst three groups: (i) cases positive for the specific pathogen; (ii) cases positive forother pathogens; and (iii) cases with negative culture. Results:    Salmonella   was associated with significantlyhigher body temperatures at presentation. Vibrio parahaemolyticus   ( VP  ) was associated with a significantlyshorter duration of diarrhoea and of abdominal pain at presentation. Other variables were not helpful inpredicting the type of stool pathogen. Conclusion:   In patients presenting with acute infectious diarrhoea inan A&E setting in Hong Kong, Salmonella   and VP   may be suspected according to the clinical featuresidentified in this study. (  Hong Kong  2003;10:162-168) Keywords : Emergency department, gastroenteritis, pathogens, predicting factors, stool culture Correspondence to:Chan Siu Wa, Stewart,  MBBS(Syd), FRCSEd(A&E), FHKAM(EmergencyMedicine) Prince of Wales Hospital , Accident and Emergency Department,30-32 Ngan Shing Street, Shatin, N.T., Hong Kong  Email: Cheung Wai Lun,  FRCSEd, FHKAM(Emergency Medicine), FHKAM(Community Medicine) Prince of Wales Hospital , Department of Microbiology,30-32 Ngan Shing Street, Shatin, N.T., Hong KongNg King Cheung,  MBChB, FRCPA, FHKAM(Pathology) Donald J Lyon,  MBChB, FHKAM(Pathology), FRCPath The Chinese University of Hong Kong ,   Centre for Clinical Trialsand Epidemiological Research, Faculty of Medicine, Shatin, N.T.,Hong KongLam Kwok Wai, Peggo,  M Phil The Chinese University of Hong Kong ,   Accident & EmergencyMedicine Academic Unit, Shatin, N.T., Hong KongTimothy H Rainer,  MBChB, MRCP, MD Introduction Infectious diarrhoea is commonly encountered inemergency departments, and the causes can be viral,bacterial or protozoan. With stool cultures, it usuallytakes 3-5 days to identify or exclude bacterialpathogens. Diarrhoea of non-infectious causes mayalso mimic infectious diarrhoea in its initialpresentation. We postulated that it might be possibleto identify, or speculate with evidence-basedconviction, certain bacterial pathogens responsible forinfectious diarrhoea, based on the presenting clinicalfeatures of patients as they attended the emergencydepartment (ED). The objective of this investigationwas to explore the possibility of correlating presentingclinical and demographic features with the specifictype of stool pathogen. The benefits and practicalimplications of predicting with good confidencecertain pathogens, simply on clinical grounds, include  Chan et al./Acute infectious diarrhoea in adults  163 (i) prompt measures in food-poisoning investigationsand public health precautions, (ii) guidance as to thenecessity for stool culture, and (iii) guidance in theindication and the choice of empiric antibiotictreatment. Materials and methods A retrospective study was performed in the setting of the Accident & Emergency (A&E) department in anurban acute-care hospital in Hong Kong, with anannual patient attendance of 190,000. The inclusioncriteria were adults (age ≥ 16); presentation to the A&Edepartment with clinical features suggestive of acuteinfectious diarrhoea; treated as outpatients with orwithout a period of observation; and with stoolcultures requested from A&E. The study period wasfrom 1 January 1999 to 31 December 1999. Laboratory methods  Stool samples were collected into sterile containers andsent for investigation according to standard laboratoryprocedures. 1   Salmonella species (spp.), Shigella spp.,Campylobacter spp.  (thermophilic group), and Vibrio spp. , were routinely looked for by inoculation of thestool samples onto xylose lysine desoxycholate agar(XLD) plate, Selenite-F (SF) enrichment broth [withsubsequent subculture onto desoxycholate citrate agar(DC) plate], thiosulphate citrate bile sucrose (TCBS)agar plate as well as Skirrow's agar plate. The XLD,SF, DC and TCBS plates were incubated at 37°Caerobically for 18 to 24 hours while the Skirrow'splates were incubated under microaerophilicconditions at 42°C for 48 hours. Suspected bacterialcolonies of primary stool pathogens were picked forfurther identification by biochemical and serologicalmethods, and antibiotic susceptibilities were alsoperformed. For suspected cases of cholera or waterystool samples, additional media such as Monsur'sagar plate and alkaline peptone water were also usedto enhance the isolation of V. cholerae. Aeromonas spp.  and Plesiomonas spp.  were occasionally foundduring the routine search for Salmonella spp. and Shigella spp. Inclusion of cases  Our subjects were derived from two groups of patients:Group A, consisting of cases with positive stoolculture, and Group B, consisting of cases with negativestool culture – the control group. Group A cases wereobtained by tracing all positive stool cultures requestedfrom the A&E department during the study period,reviewing the relevant clinical records, and includingevery case meeting the inclusion criteria above. GroupA cases were then further divided into six sub-groups,according to the specific type of pathogen identified.Group B (control) patients were obtained by firsttracing all stool cultures of adult patients (age ≥ 16)requested from the A&E department, during the studyperiod, that were negative. A random sample of 150cases was then generated. From this sample, a finalgroup of patients meeting inclusion criteria wasformed. The decision for inclusion or exclusion of eachcase was determined by an emergency physician (SC)after retrospective review of the relevant clinical notes.Cases with diarrhoea duration longer than 14 days atpresentation were considered non-acute, and wereexcluded.Demographic, clinical and laboratory details of thesubjects were retrieved and collated into a database.The demographic variables evaluated were age andmonth/season of presentation. The clinical variablesat presentation to A&E included: body temperature,duration of diarrhoea, number of loose stools per day,presence of bloody diarrhoea, the presence andduration of abdominal pain and vomiting, therequirement of intravenous fluid and observation wardutilisation. The definition of each clinical variable isgiven in Table 1.Statistical analyses were performed to look at theclinical and demographic features of each pathogenin comparison with those of other pathogens, and incomparison with those of the control group. Factorsmeasured in continuous scale were analysed by one-way ANOVA, followed by post-hoc tests. Factorsmeasured in categorical scale were analysed with cross-tabulation tables using chi-squared test or Fisher'sexact test as appropriate.  Hong Kong j. emerg. med.      Vol. 10(3)      Jul 2003 164 Results Group A consisted of 130 patients, subdivided intosix groups according to the specific pathogen isolated.Pathogens identified were Vibrio parahaemolyticus  (n=55), Samonella spp.  (n=45), Plesiomonas spp. (n=12), Campylobacter spp.  (n=9), Aeromonas spp. (n=9), and Shigella spp.  (n=8). (Eight patients hadtwo pathogens isolated). Group B consisted of 119patients with negative stool culture. Laboratory dataover the study period showed that there was a total of 524 negative stool culture results (including repeattests), from which a random sample of 150 wasgenerated. Out of these 150 cases, the clinical recordsof five (3.3%) cases were not available for review. Fromthe remaining 145 cases, 26 were excluded, resultingin a final number of 119 cases. The main reason forexclusion of these 26 cases was that the patients didnot present with a clinical picture of acute infectiousdiarrhoea. Amongst those excluded were cases withduration of diarrhoea longer than 14 days, cases thatrequired repeat stool culture requests from follow-upvisits; stool culture requested as routine health checks;and stool culture requested due to suspicion of food-poisoning (through history of exposure), but withoutsymptoms from the patients.Table 2 shows the number and proportion of patientswith each type of pathogen, and within each respectivestatistical group. 'Negative' represents the culture- Table 1.  Definition of clinical parameters obtained from retrospective record review Highest body temperature.  The highest oral temperature recorded during attendance in A&E or observation unit stay. Bloody diarrhoea.  Either suspected by history from patient, or confirmed by inspection of stool, and not due to local anorectalbleeding. Duration of diarrhoea (days).  The duration of diarrhoea from onset to the completion of stay in A&E or observation unit.(0-24 hours=1 day; 24-48 hours=2 days, 48-72 hours=3 days, etc.) Number of loose stools per day.  Can be by history (maximum number per day) on presentation, or recorded in observation unitcharts, whichever is larger in number. Duration of abdominal pain (days).  The duration of abdominal pain from onset to the completion of stay in A&E or observationunit. (0-24 hours=1 day; 24-48 hours=2 days; 48-72 hours=3 days, etc.) Duration of vomiting (days).  The duration of vomiting from onset to the completion of stay in A&E or observation unit. (0-24hours=1 day; 24-48 hours=2 days, 48-72 hours=3 days, etc.) Intravenous fluid requirement.  Dehydration or poor oral intake requiring intravenous fluid therapy. Table 2.  Number of patients with each type of pathogen and within each statistical group NegativePositiveOthers Aeromonas  N1199121%47.8%3.6%48.6% Campylobacter  N1199121%47.8%3.6%48.6% Plesiomonas  N11912118%47.8%4.8%47.4% Salmonella  N1194585%47.8%18.1%34.1% Shigella  N1198122%47.8%3.2%49.0% Vibrio parahaemolyticus  N1195575%47.8%22.1%30.1% N – number; Negative – culture-negative; Positive – culture-positive for the specific pathogen; Others – culture-positive for other pathogens  Chan et al./Acute infectious diarrhoea in adults  165 negative group; while 'Positive' represents presence of the specific pathogen, and 'Others' represents presenceof other pathogens.Table 3 shows results of the factors measured incontinuous scale, for which analyses by ANOVA haveshown significance. Overall P values are shown.Significance implied that at least one group wasdifferent from the others, and post-hoc tests were thenperformed to make pairwise comparisons amongst the Table 4.  Categorical variables associated with each pathogen, with P values showing the strength of association Month ofBloodyPresence ofPresence ofIV fluidObservationpresentationdiarrhoeaabdominal painvomitingward utilized Aeromonas  <0.0005 0.3830.1170.643 <0.0005<0.0005 Campylobacter  <0.0005 0.0970.1170.210 <0.0005<0.0005 Plesiomonas  <0.0005 0.3950.1660.512 <0.0005<0.0005 Salmonella  <0.0005 0.0570.0960.643 <0.0005<0.0005 Shigella  <0.00050.008 0.1930.319 <0.0005<0.0005 VP  <0.0005 0.5180.1620.084 <0.0005<0.0005 Month of presentation: between May-October (vs. November-April) VP – Vibrio parahaemolyticus  group. With each pathogen, there are three rowsshowing the post-hoc test results. The first rowrepresents the significance level when comparing'Negative' with 'Others'; the second row for comparing'Negative' with 'Positive'; and the third row forcomparing 'Others' with 'Positive'.Table 4 refers to factors or variables measured incategorical scale. P values show the strength of association between the pathogen and the variable. Table 3.  Identifying factors associated with stool pathogens: continuous variables that showed significance with ANOVA test  P valuePathogenGroupNMeanSDMinMedianMaxOverallPost-hoc testHighest  Salmonella  Negative10736.810.7135.9036.6039.200.001 temperature Others8037.260.8336.0037.2039.20<0.0005<0.0005 recorded Positive4337.861.1036.0038.0040.000.007 VP  Negative10736.810.7135.9036.6039.20<0.005Others7237.601.0336.0037.4540.00<0.0050.003Positive5137.280.8736.0037.2039.000.201 Duration of   VP  Negative1162.973.051.002.0014.000.944 diarrhoea (days) Others742.762.851.002.0014.000.002<0.0005Positive531.341.861.001.0014.000.003 No. of loose stools  VP  Negative1136.874.231.006.0020.000.107 per day Others658.495.411.0010.0020.00<0.0005< .0005Positive5210.215.701.0010.0030.000.188 Duration of   VP  Negative1193.003.640.001.0021.000.363 abdominal pain Others732.372.250.002.0014.00<0.0005<0.0005 (days) Positive531.040.680.001.005.00<0.0005 Plesiomonas  Negative1193.003.640.001.0021.000.010Others1151.871.940.001.0014.000.0050.001Positive111.180.980. Note: Some cases had missing data on one or more clinical features VP – Vibrio parahaemolyticus  ; Negative – culture-negative; Positive – culture-positive for the specific pathogen; Others – culture-positive for otherpathogens; Post-hoc tests, significance levels: Row 1 – comparing "Negative" with "Others", Row 2 – comparing "Negative" with "Positive", Row3 – comparing "Others" with "Positive"  Hong Kong j. emerg. med.      Vol. 10(3)      Jul 2003 166 Relevant findings are summarised below:1.  Salmonella   was associated with significantly higherbody temperatures compared to the culture-negative group, and to other pathogens.2.  Vibrio parahaemolyticus (VP)   was associated witha significantly shorter duration of diarrhoea andof abdominal pain at presentation, compared toother pathogens and to the culture-negative group.3.The mean number of loose stools per day atpresentation was significantly higher in patients with VP  , compared to culture-negative cases (10.21 vs.6.87, P<0.005), but was not significantly higher thanthat of other pathogens (10.21 vs. 8.49, P=0.188).4.The mean highest temperature recorded for patientswith VP   was significantly higher than that of theculture-negative group (37.28°C vs. 36.81°C,P=0.003), but there was no significant differencecompared to that of other pathogens (37.28°C vs.37.60°C, P=0.201).5.The mean duration of abdominal pain atpresentation was significantly shorter in patientswith Plesiomonas  , compared to culture-negativecases (1.18 vs. 3.00 days, P=0.001), but was notsignificantly shorter than that of other pathogens(1.18 vs. 1.87 days, P=0.17).6.None of the stool pathogens, except Shigella  , wassignificantly associated with bloody diarrhoea.7.Compared with the control-group, all sixpathogens were individually significantly associatedwith the variables of month of presentation (May-October), requirement of intravenous fluid, andobservation ward utilization. These variables werenot useful in distinguishing the type of pathogen.8.The variables of age, presence of vomiting,duration of vomiting at presentation, and presenceof abdominal pain were not useful in distinguishingthe type of pathogen. Discussion An earlier study had identified and reported predictingfactors of positive stool culture in adult patients inthe ED setting. 2  This investigation was a furtherattempt to correlate presenting patient variables withspecific pathogen type.Most cases of infectious diarrhoea are self-limitingwithin several days, and the mainstay of therapy issymptomatic relief and maintenance of hydration. 3,4 Most authorities would consider a first-handspeculation of the responsible pathogen to be non-essential to patient management. Nevertheless, thereare reasonable grounds for this study. It would beuseful if the current practice in the management of community-acquired pneumonia is considered.Empiric antibiotic treatment and risk classification of patients are based on the clinical presentation, which,in turn, predicts the likelihood of certain types of pathogens. 4-8  Likewise, depending on the clinicalpresentation, empiric antibiotic treatment of infectious diarrhoea is currently recommended for asubset of adult patients with moderate to severesymptoms (e.g. with fever, bloody diarrhoea, ordehydration). 3,4,9-12  The Infectious Diseases Society of America guidelines for empiric antibiotic treatmentof infectious diarrhoea, published in 2001, evenspecifically recommended quinolones for suspectedshigellosis in adults, macrolides for suspected resistant Campylobacter  , and the avoidance of anti-motilityagents if shiga toxin-producing Escherichia coli O157: H7   was suspected. 9  Hence, in this connection, wepostulated that the results of our study could be usefulto guide the choice of empiric antibiotic treatmentfor patients with acute infectious diarrhoea. A previousstudy involving the same database had comparedpresenting features according to pathogen isolated,amongst culture-positive patients only. 13,14  In thepresent study, the culture-negative control-group wasfurther incorporated into the analyses, so as to achieveour stated objective.The issue of Campylobacter   antimicrobial resistance isparticularly relevant locally in Hong Kong. In as muchas 77.8% of adult A&E patients with Campylobacter  gastroenteritis, the isolate was resistant to ciprofloxacin,and all isolates were sensitive to erythromycin. 13,14 There has been a steady and alarming increase inquinolone-resistant Campylobacter   infectionsworldwide. 9,15-17  Identifying clinical features of  Campylobacter   infection would therefore be importantto guide the choice of antibiotic for empiric treatment.Factors associated with Campylobacter   infection were
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!