Divergent recovery trajectories of intertidal and subtidal coral communities highlight habitat-specific recovery dynamics following bleaching in an extreme macrotidal reef environment.

Coral reefs face an uncertain future punctuated by recurring climate-induced disturbances. Understanding how reefs can recover from and reassemble after mass bleaching events is therefore important to predict their responses and persistence in a rapidly changing ocean. On naturally extreme reefs characterized by strong daily temperature variability, coral heat tolerance can vary significantly over small spatial gradients but it remains poorly understood how this impacts bleaching resilience and recovery dynamics, despite their importance as resilience hotspots and potential refugia. In the macrotidal Kimberley region in NW Australia, the 2016 global mass bleaching event had a strong habitat-specific impact on intertidal and subtidal coral communities at our study site: corals in the thermally variable intertidal bleached less severely and recovered within six months, while 68% of corals in the moderately variable subtidal died. We therefore conducted benthic surveys 3.5 years after the bleaching event to determine potential changes in benthic cover and coral community composition. In the subtidal, we documented substantial increases in algal cover and live coral cover had not fully recovered to pre-bleaching levels. Furthermore, the subtidal coral community shifted from being dominated by branching Acropora corals with a competitive life history strategy to opportunistic, weedy Pocillopora corals which likely has implications for the functioning and stress resilience of this novel coral community. In contrast, no shifts in algal and live coral cover or coral community composition occurred in the intertidal. These findings demonstrate that differences in coral heat tolerance across small spatial scales can have large consequences for bleaching resilience and that spatial patchiness in recovery trajectories and community reassembly after bleaching might be a common feature on thermally variable reefs. Our findings further confirm that reefs adapted to high daily temperature variability play a key role as resilience hotspots under current climate conditions, but their ability to do so may be limited under intensifying ocean warming.

Management of post-splenectomy patients in the Netherlands.

After splenectomy, patients are at increased risk of sepsis with considerable mortality. The risk of sepsis can be reduced by immunising these patients and by prescribing antibiotic prophylaxis. The purpose of our study was to determine compliance with the international standards for the management of splenectomised patients in the Netherlands by investigating: (i) vaccination rates, (ii) the prescription of antibiotics and (iii) information in discharge letters. A retrospective review of the medical records and discharge correspondence of 609 splenectomy patients from 1997 to 2008 was performed. Data were collected from 28 hospitals. Adherence to vaccination guidelines and the prescription of antibiotics were assessed. It was found that 85.4% of post-splenectomy patients received pneumococcal vaccination, 39.4% received Haemophilus influenzae type B and 32.3% received meningococcal group C vaccination. Also, 12.4% of patients were discharged on prophylactic antibiotics. In less than 25% of cases were adequate recommendations regarding post-splenectomy management given to the general practitioner (GP). In the Netherlands, compliance with recommendations for the management of patients after splenectomy is insufficient. Fifteen percent of patients do not receive vaccination against pneumococci and the majority of patients do not receive antibiotic prophylaxis. The development and implementation of a national guideline for splenectomised patients is urgently required.

Assessment of splenic function.

Hyposplenic patients are at risk of overwhelming post-splenectomy infection (OPSI), which carries mortality of up to 70%. Therefore, preventive measures are warranted. However, patients with diminished splenic function are difficult to identify. In this review we discuss immunological, haematological and scintigraphic parameters that can be used to measure splenic function. IgM memory B cells are a potential parameter for assessing splenic function; however, more studies are necessary for its validation. Detection of Howell-Jolly bodies does not reflect splenic function accurately, whereas determining the percentage of pitted erythrocytes is a well-evaluated method and seems a good first-line investigation for assessing splenic function. When assessing spleen function, (99m)Tc-labelled, heat-altered, autologous erythrocyte scintigraphy with multimodality single photon emission computed tomography (SPECT)-CT technology is the best approach, as all facets of splenic function are evaluated. In conclusion, although scintigraphic methods are most reliable, they are not suitable for screening large populations. We therefore recommend using the percentage of pitted erythrocytes, albeit suboptimal, as a first-line investigation and subsequently confirming abnormal readings by means of scintigraphy. More studies evaluating the value of potentially new markers are needed.

Struggling with recurrent Clostridium difficile infections: is donor faeces the solution?

Patients with recurrent Clostridium difficile infections (CDI) in hospitals and the community constitute an increasing treatment problem. While most patients with a first infection respond to either metronidazole or oral vancomycin, therapy in recurrent C. difficile infections tends to fail repeatedly. Lack of alternative treatment options can be a tremendous burden, both to patients and their treating physicians. Most guidelines recommend prolonged oral vancomycin pulse and or tapering schedules, but evidence-based treatment strategies are lacking. The role of immunoglobulins, whey prepared from vaccinated cows, probiotics or other antibiotics is unclear. Since 1958 several case series and case reports describe a treatment strategy where faecal infusions are successfully given for the treatment of recurrent CDI. Restoring intestinal flora has been historically thought of as the mechanism responsible for cure in these patients. In the literature, more than 150 patients have received faeces from a healthy donor, either infused through an enema, or through a nasoduodenal or nasogastric tube. We summarise the literature regarding treatment with donor faeces for recurrent CDI, and introduce the FECAL trial, currently open for inclusion.

Short-course antibiotic treatment in acute exacerbations of chronic bronchitis and COPD: a meta-analysis of double-blind studies.

BACKGROUND: A study was undertaken to determine whether a short course of antibiotic treatment (< or = 5 days) is as effective as the conventional longer treatment in acute exacerbations of chronic bronchitis and chronic obstructive pulmonary disease (COPD). METHODS: MEDLINE, EMBASE and the Cochrane central register of controlled trials were searched to July 2006. Studies considered eligible were double-blind randomised clinical trials including adult patients > or = 18 years of age with a clinical diagnosis of exacerbation of COPD or chronic bronchitis, no antimicrobial therapy at the time of diagnosis and random assignment to antibiotic treatment for < or = 5 days versus > 5 days. The primary outcome measure was clinical cure at early follow-up on an intention-to-treat basis. RESULTS: 21 studies with a total of 10 698 patients were included. The average quality of the studies was high: the mean (SD) Jadad score was 3.9 (0.9). At early follow-up (< 25 days), the summary odds ratio (OR) for clinical cure with short treatment versus conventional treatment was 0.99 (95% CI 0.90 to 1.08). At late follow-up the summary OR was 1.0 (95% CI 0.91 to 1.10) and the summary OR for bacteriological cure was 1.05 (95% CI 0.87 to 1.26). Similar summary ORs were observed for early cure in trials with the same antibiotic in both arms and in studies grouped by the antibiotic class used in the short-course arm. CONCLUSIONS: A short course of antibiotic treatment is as effective as the traditional longer treatment in patients with mild to moderate exacerbations of chronic bronchitis and COPD.

[Treatment of recurrent Clostridium difficile-associated diarrhoea with a suspension of donor faeces]. / Behandeling van recidiverende Clostridium dificile-geassocieerde diarree met een suspensie van donorfeces.

OBJECTIVE: To study the effect of treating recurrent Clostridium difficile-associated diarrhoea (CDAD) with a suspension of donor faeces. DESIGN: Uncontrolled interventional study. METHOD: Patients that, despite adequate antibiotic therapy, had developed at least 2 recurrences ofCDAD, including at least one recurrence that had been treated with a vancomycin tapering regimen, were included in the study. Relatives or volunteers served as faeces donor. All donors were previously examined for the presence of HIV, hepatitis B- and C-virus, and acute infection with cytomegalovirus or Epstein-Barr virus. The donor faeces were examined for the presence of C. difficile, Yersinia, Campylobacter, Shigella, Salmonella, and parasites. Before the infusion of donor faeces, the patients were treated for 4 days with vancomycin 500 mg q.i.d., followed by colon lavage. The suspension of 150 g of donor faeces dissolved in 300-400 ml of NaCl was infused into the jejunum via a duodenal catheter or into the caecum via colonoscopy. RESULTS: 7 CDAD patients were included and treated, including 2 with the hypervirulent C. difficile-strain PCR ribotype 027, toxinotype III. In 5 patients, the defaecation frequency returned to normal almost immediately after treatment and the cultures and toxin tests for C. difficile were repeatedly negative. In the remaining 2 patients, the treatment was successful after a repeated infusion of faeces from a different donor. CONCLUSION: Treatment with donor faeces seems promising for patients who develop repeated recurrences despite adequate therapy and could be valuable in the future during (local) epidemics of the PCR ribotype 027 strain. A randomised nationwide study (FECAL trial) has been started in order to determine the efficacy of this treatment.

Three vs. 10 days of amoxycillin-clavulanic acid for type 1 acute exacerbations of chronic obstructive pulmonary disease: a randomised, double-blind study.

The optimal duration of antibiotic treatment for acute exacerbations of chronic obstructive pulmonary disease (AECOPD) is unknown. This study compared the outcome of treatment for 3 vs. 10 days with amoxycillin-clavulanic acid of hospitalised patients with AECOPD who had improved substantially after initial therapy for 3 days. Between November 2000 and December 2003, 56 patients with AECOPD were enrolled in the study. Unfortunately, because of the low inclusion rate, the trial was discontinued prematurely. Patients were treated with oral or intravenous amoxycillin-clavulanic acid. Patients who showed improvement after 72 h were randomised to receive oral amoxycillin-clavulanic acid 625 mg or placebo, four times daily for 7 days. The primary outcome measure of the study was clinical cure after 3 weeks and 3 months. Of 46 patients included in the final analysis, 21 were in the 3-day treatment group and 25 were in the 10-day treatment group. After 3 weeks, 16 (76%) of 21 patients in the 3-day treatment group were cured, compared with 20 (80%) of 25 in the 10-day treatment group (difference -3.8%; 95% CI -28 to 20). After 3 months, 13 (62%) of 21 patients were cured, compared with 14 (56%) of 25 (difference 5.9%; 95% CI -23 to 34). Microbiological success, symptom recovery, the use of corticosteroids, the duration of oxygen therapy and the length of hospital stay were comparable for both treatment groups. It was concluded that 3-day treatment with amoxycillin-clavulanic acid can be a safe and effective alternative to the standard 10-day treatment for hospitalised patients with AECOPD who have improved after initial therapy for 3 days.

[Adverse events following influenza vaccination: reaction to specific reports and the necessity of a central registration system]. / Bijwerkingen na influenzavaccinatie: reactie op enkele incidenten en het belang van centrale registratie.

The influenza vaccine is considered safe, but information on vaccine-related adverse events is limited and a nationwide overview of adverse events is lacking. In 2006, after deaths occurred in Israel and the Netherlands following influenza vaccination, the Dutch Ministry of Health, Welfare and Sport (VWS) asked the National Institute for Public Health and the Environment (RIVM) twice for a recommendation regarding the continuation of the national vaccination campaign. After 4 deaths were reported in Israel in October 2006 following administration of Vaxigrip, the Dutch vaccination campaign was suspended for one week. One month later, 4 additional deaths were reported after influenza vaccination in the Netherlands. The newly appointed outbreak management team concluded that a causal relationship between vaccination and the deaths was highly unlikely, based on data regarding the individual cases, background mortality rates and prior reports of adverse events. Further suspension of the vaccination campaign was deemed unnecessary this time. A centralised nationwide registry of adverse events has since been established to provide further insight into the incidence of adverse events following influenza vaccination. Physicians are advised to report potential adverse events following influenza vaccination to the Netherlands Pharmacovigilance Centre Lareb (www.lareb.nl).

[An outbreak of Q fever in The Netherlands--possible link to goats]. / Een uitbraak van Q-koorts in Nederland--mogelijk verband met geiten.

In 2007, 73 cases of Q fever were identified through reports and retrospective analyses; the affected region extended from Tilburg in the southwest to Arnhem in the northeast. The infections arose in late spring, particularly in May and June. Several spontaneous abortions due to Q fever occurred on 4 dairy goat farms in the same region. The national incidence of spontaneous abortion due to Q fever was 6 cases in 2006 and 7 in 2007. Climatically, this southern region was extraordinarily dry during April 2007. All pregnant women from a small region with the highest incidence in northeast North Brabant were called for diagnostic testing. Infected patients were followed for symptoms and ultrasound was performed as indicated. A definitive source of the infection could not yet be identified. Favourable climatic conditions were suspected as the cause for the combination of widespread dissemination among goats and transmission to humans. Q fever is a zoonosis caused by Coxiella burnetti, a microorganism dispersed in great numbers in the area in which an infected animal gives birth. C. burnetti is particularly resistant to chemical and physical factors and can disperse by air across large distances under dry climatic conditions. Q fever should be considered in patients in The Netherlands who present with lower airway infection and, in rare cases, hepatitis. Reporting atypical clusters ofpneumonia to the Municipal Health Service (GGD) is advisable. The GGD maintains close contact with Animal Health Services, which is aware of current infectious animal diseases. Targeted investigation can identify the source of infection and eliminate it. Greater awareness can prevent delays in diagnosis and treatment and help identify chronic forms at an early stage or prevent them.

[Severe phenytoin intoxication in patients with hypoalbuminaemia]. / Ernstige fenytoïne-intoxicatie bij patiënten met hypoalbuminemie.

Two patients, a 35-year-old woman and a 60-year-old man, developed severe neurological side effects during treatment with phenytoin: disorientation, myoclonia, hallucinations and drowsiness in the first patient and a comatose state in the second. The woman had spina bifida, a ventriculoperitoneal drain because of hydrocephalus, recurrent urinary-tract infection, and a history of status epilepticus. The man suffered from diabetic ketoacidosis complicated by epileptic convulsions. In both patients, the total phenytoin concentration in the blood was within the therapeutic range of 8-20 mg/l. However, both had low serum albumin concentrations, below 25 g/l. Low serum albumin levels are associated with increased concentrations of the free fraction of phenytoin. Toxic levels of free phenytoin were found: 4 and 8 mg/l, respectively, while the therapeutic range is 0.5-2 mg/l. The first patient recovered after treatment with phenytoin was stopped, after which she was placed on a lower dosage; the second patient died. When prescribing phenytoin to patients with hypoalbuminaemia, one should be aware of the risk of intoxication due to a high level of free phenytoin and consequently an increased risk of severe neurological side effects.

[Optimising antibiotic policies in the Netherlands. IX. SWAB guidelines for antimicrobial therapy in adults with acute infectious diarrhoea]. / Optimaliseren van het antibioticabeleid in Nederland. IX. SWAB-richtlijn voor antimicrobiële therapie bij acute infectieuze diarree.

The 'Stichting Werkgroep Antibioticabeleid' (SWAB; Dutch Working Party on Antibiotic Policy) develops evidence-based guidelines for the use of antibiotics in hospitalised adults. This guideline on acute infectious diarrhoea (AID) concerns the antibiotic treatment of acute infectious inflammation of the gastrointestinal tract, manifesting primarily as diarrhoea. AID can be subdivided into community-acquired diarrhoea, traveller's diarrhoea and hospital-acquired (nosocomial) diarrhoea. In the first 2 categories, the need for antibiotic treatment is generally restricted to individuals with severe illness, dysentery or a predisposition to complications. High rates of primary fluoroquinolone resistance can be found in human Campylobacter isolates from the Netherlands and from other parts of the world. Therefore, if antibiotic treatment is necessary for community-acquired AID or AID in travellers returning to the Netherlands, it is advised to use oral azithromycin for 3 days as empirical treatment. If intravenous treatment is necessary, the combination of ciprofloxacin and erythromycin for 5-7 days may be considered. As soon as the identity of the causative organism is known, antimicrobial treatment should be tailored accordingly.

[Two adult patients with pneumococcal peritonitis]. / Twee volwassen patiënten met pneumokokkenperitonitis.

Two women, aged 31 and 37 years, had abdominal pain and fever several months after giving birth and a few weeks after receiving an intrauterine device. Both patients were admitted and treated under the working diagnosis of pelvic inflammatory disease (PID). They appeared to have pneumococcal adnexitis and pneumococcal peritonitis. Both patients recovered after initiating directed antibiotic treatment. Peritonitis in previously healthy adults is seldom caused by pneumococci. Standard antibiotics that are effective when given empirically for PID may be a suboptimal treatment for pneumococcal peritonitis.

[Life-threatening infections with a new strain of Clostridium difficile]. / Levensbedreigende infecties met een nieuwe variant van Clostridium difficile.

Three men, aged 39, 73, and 66 years, respectively, developed an infection with a new strain ofClostridium difficile, ribotype 027.C.difficile-associated diarrhoea (CDAD) occurred in two patients after multiple abdominal surgery and in the third patient one week after autologous haematopoietic cell transplantation. Within a few days, despite antibiotic therapy, all three patients developed severe (pseudomembranous) colitis with sepsis for which admission to the Intensive Care Unit was required. Two patients underwent (sub)total colectomy and received an intensive course of oral and/or rectal vancomycin. In all patients who develop diarrhoea in hospital, especially during or after treatment with antibiotics or chemotherapeutic agents, an infection with C. difficile ribotype 027 should be suspected. Recent outbreaks of this hypervirulent strain of C. difficile have been reported in Canada, the United States, United Kingdom, and The Netherlands. Demonstration of C. difficile toxin in faeces confirms the clinical suspicion of CDAD and ribotyping of the strain may reveal whether the 027 strain is present. For treatment of these 027 infections, vancomycin is preferred to metronidazole. After a severe course of colitis or in case of recurrence a 'tapering and pulse' course ofvancomycin can be prescribed; alternatively, treatment with bovine antibody-enriched whey may be considered. The introduction of this hypervirulent strain has led to reinforcement of the hygienic measures in accordance with the recommendations of the Dutch Working Party on Infection Prevention and a policy to deter the use of fluoroquinolones.

Mycobacterium xenopi in HIV-infected patients: an emerging pathogen.

BACKGROUND: Mycobacterium xenopi is associated with pulmonary disease in patients with loss of local or general host defence. OBJECTIVES: To determine the occurrence of M. xenopi in our hospital during 1987-1992 and 1993-1996, as well as the association of M. xenopi with HIV infection in 1993-1996; to evaluate the clinical significance of M. xenopi in HIV-seropositive patients. DESIGN: Retrospective review of charts and classification of patients based on earlier definitions derived from the American Thoracic Society. SETTING: Tertiary hospital. PATIENTS: Patients with a positive isolate of M. xenopi from January 1987 until December 1996. MAIN OUTCOME MEASURES: During 1993 1996, a significant increase in the number of patients with M. xenopi was found compared with 1987-1992. Of 25 patients, 22 were HIV-seropositive. RESULTS: The HIV-seropositive patients were classified as having definite (n = 5), probable (n = 9) and unlikely disease (n = 8) due to M. xenopi. Symptoms, median CD4 cell count, treatment and outcome did not differ between these groups. CONCLUSIONS: M. xenopi is an emerging pathogen, especially in HIV-infected patients. The criteria of the American Thoracic Society for disease due to non-tuberculous mycobacteria do not seem applicable to M. xenopi in HIV-infected patients. We suggest that two positive cultures of M. xenopi and no other likely cause of symptoms present should be considered the criteria for diagnosis of M. xenopi disease in HIV-infected patients.

Circadian variations in serum levels and the renal toxicity of aminoglycosides in patients.

Animals show a faster clearance and a lower incidence of nephrotoxicity and ototoxicity when aminoglycosides are administered during the activity period. Human data on a circadian rhythm in pharmacokinetics are conflicting, and there are no data on a circadian rhythm in toxicity. When aminoglycosides are administered once daily, as is often done, a circadian rhythm in pharmacokinetics or toxicity could have clinical implications. In a prospective study we investigated the influence of drug administration time on serum drug levels and the incidence of nephrotoxicity in 221 patients with serious infections treated with gentamicin or tobramycin once daily. We did not find statistically significant differences in trough or peak levels for the three time periods (midnight to 7:30 AM, 8 AM to 3:30 PM, and 4 to 11:30 PM). Nephrotoxicity occurred significantly more frequently when the aminoglycosides were administered during the rest period (midnight to 7:30 AM; p = 0.004). In addition to the coadministration of high-dose furosemide or other nephrotoxic antibiotics and the duration of treatment, the time of administration was still an independent risk factor in a multivariate analysis.

Guidelines for adults on self-medication for the treatment of acute diarrhoea.

Acute uncomplicated diarrhoea is commonly treated by self-medication. Guidelines for treatment exist, but are inconsistent, sometimes contradictory, and often owe more to dogma than evidence. An ad hoc multidisciplinary group has reviewed the literature to determine best practice. In general it is recognized that treatment of acute episodes relieves discomfort and social dysfunction. There is no evidence that it prolongs the illness. Self-medication in otherwise healthy adults is safe. Oral loperamide is the treatment of choice. Older anti-diarrhoeal drugs are also effective in the relief of symptoms but carry the risk of unwanted adverse effects. Oral rehydration solutions do not relieve diarrhoea, and confer no added benefit for adults who can maintain their fluid intake. Probiotic agents are, at present, limited in efficacy and availability. Antimicrobial drugs, available without prescription in some countries, are not generally appropriate for self-medication, except for travellers on the basis of medical advice prior to departure. Medical intervention is recommended for the management of acute diarrhoea in the frail, the elderly (> 75 years), persons with concurrent chronic disease, and children. Medical intervention is also required when there is no abatement of the symptoms after 48 h, or when there is evidence of deterioration such as dehydration, abdominal distension, or the onset of dysentery (pyrexia > 38.5 degrees C and/or bloody stools).

Protozoal enteric infections among expatriates in Bangladesh.

In order to study the prevalence, incidence, and symptoms of infections with Giardia lamblia and Entamoeba histolytica, we followed 251 expatriates in Bangladesh over a 1-year period. Microscopic examination of fecal specimens was performed upon enrollment, at 3-month intervals, and during episodes of diarrhea. Specimens were cultured for bacteria and samples of serum and saliva were collected for antibody studies (IgG and SIgA). The prevalence of G. lamblia infections was 5.2% and the incidence 11.8%. Children aged less than or equal to 10 years and newcomers were most frequently infected (P less than 0.02). Symptoms were present in 37% of the subjects infected with G. lamblia. A systemic antibody response was observed in 57% of symptomatic patients and 35% of asymptomatic subjects during the first 2 months of infection. The prevalence and the incidence of E. histolytica infection were 3.2% and 8.6%, respectively. Infections with E. histolytica were correlated with the duration of stay in Bangladesh; less than 1 year 7% vs. greater than 3 years 26% (P less than 0.01). Most expatriates infected with E. histolytica were asymptomatic (90%). Three adult patients, who were resident in Bangladesh for less than 1 year, were symptomatic but none of them developed dysentery or a serological response. Four of 25 asymptomatic subjects had significant antibody titers. Three of these people were seropositive for ameba at the beginning of the study. The local immune response, reflected by specific secretory IgA in saliva samples, correlated poorly with both E. histolytica and G. lamblia infections.

Known bacterial virulence factors do not explain the variation in urinary cytokine levels in patients with urosepsis.

We measured urinary endotoxin, IL-6 and IL-8 levels in 23 patients with gram-negative urosepsis. The endotoxin and cytokine levels showed a 100-1000 fold range. No correlation was found between levels of urinary endotoxin, and IL-6 or IL-8 levels. In all cases bacterial numbers were > or = 10(5) CFU ml-1 urine. The endotoxin content of the isolated microorganisms neither correlated with the urinary cytokine levels, nor with IL-6 and IL-8 levels obtained in vitro when 10(3) log-phase CFU of each of the bacteria were incubated with heparinized whole blood of three healthy donors. Neither the haemolysin phenotype of the bacteria, nor the presence of the P-pili gene was correlated with the cytokine response in vivo or in vitro. Other factors than known bacterial virulence factors apparently contribute to the wide variation in urinary cytokine levels in urinary tract infection.

The effect of pseudomonas exotoxin A on cytokine production in whole blood exposed to Pseudomonas aeruginosa.

To determine the effect of Pseudomonas aeruginosa exotoxin A (P-ExA) on cytokine production, we studied cytokine release induced by heat-killed P. aeruginosa (HKPA) in human whole blood in the presence or absence of P-ExA. P-ExA (0.01-1 microgram ml(-1)) caused a dose-dependent decrease in HKPA-induced production of tumor necrosis factor alpha (TNF), interleukin (IL-) 10, IL-6 and IL-8 (all P<0.05). P-ExA-induced inhibition of IL-10, IL-6 and IL-8 release was not dependent on reduced TNF concentrations, since the relative attenuation of the production of these cytokines was similar in the presence or absence of a neutralizing anti-TNF antibody. The effect of P-ExA on cytokine production may offer a disadvantage to the host with respect to clearance of the infection.

Differential clinical features and stool findings in shigellosis and amoebic dysentery.

To obtain information that could assist the clinician to differentiate between shigellosis and amoebic dysentery, we compared clinical features and stool findings in 58 adult male patients in Bangladesh. Mean values indicated that patients with invasive amoebiasis were older and had a longer prehospital illness, a lower body weight, less frequent fever, a lower haematocrit and a higher white blood cell count than patients with shigellosis. The mean number of faecal leucocytes per mm3 was significantly higher in shigellosis than in amoebiasis (28,700 vs 10,300) and correlated with the estimated number of faecal leucocytes per microscopic high power field in a wet mount preparation. Patients with shigellosis more often had over 50 white blood cells per high power field. Although the mean stool pH in amoebiasis was lower than in shigellosis (6.26 vs 6.60), the difference was not statistically significant. Concentrations of stool electrolytes did not differ between the two diseases. These findings indicate that age, duration of illness, the presence of fever and the number of faecal leucocytes may help to differentiate between shigellosis and amoebic dysentery.