Serious altitude illness in travelers who visited a pre-travel clinic.

BACKGROUND: Few data are available on the incidence and predictors of serious altitude illness in travelers who visit pre-travel clinics. Travel health consultants advise on measures to be taken in case of serious altitude illness but it is not clear if travelers adhere to these recommendations. METHODS: Visitors to six travel clinics who planned to travel to an altitude of ≥3,000 m were asked to complete a diary from the first day at 2,000 m until 3 days after reaching the maximum sleeping altitude. Serious altitude illness was defined as having symptoms of serious acute mountain sickness (AMS score ≥ 6) and/or cerebral edema and/or pulmonary edema. RESULTS: The incidence of serious altitude illness in the 401 included participants of whom 90% reached ≥4,000 m, was 35%; 23% had symptoms of serious AMS, 25% symptoms of cerebral edema, and 13% symptoms of pulmonary edema. Independent predictors were young age, the occurrence of dark urine, travel in South America or Africa, and lack of acclimatization between 1,000 and 2,500 m. Acetazolamide was brought along by 77% of the responders of whom 41% took at least one dose. Of those with serious altitude illness, 57% had taken at least one dose of acetazolamide, 20% descended below 2,500 m on the same day or the next, and 11% consulted a physician. CONCLUSIONS: Serious altitude illness was a very frequent problem in travelers who visited pre-travel clinics. Young age, dark urine, travel in South America or Africa, and lack of acclimatization nights at moderate altitude were independent predictors. Furthermore, we found that seriously ill travelers seldom followed the advice to descend and to visit a physician.

Tuberculous meningitis: do modern diagnostic tools offer better prognosis prediction?

BACKGROUND: The British Medical Research Council (BMRC) staging has been extensively used to evaluate the disease severity and establish the approximate prognosis of tuberculous meningitis. AIMS: This study aimed at analyzing the predictive accuracy for mortality and neurological sequelae of a set of clinical features, laboratory tests and imaging. METHODS: We compared the British Medical Research Council (BMRC) staging with a new scoring proposal to predict the prognosis of patients with Central Nervous System Tuberculosis. Data from Ecuador was collected. A score was built using a Spiegelhalter and Knill-Jones method and compared with BMRC staging with a ROC curve. RESULTS: A total of 213/310 patients (68.7%) were in BMRC stage II or III. Fifty-seven patients died (18.3%) and 101 (32.5%) survived with sequelae. The associated predictors were consciousness impairment (p = 0.010), motor deficit (p = 0.003), cisternal effacement (p = 0.006) and infarcts (p = 0.015). The new score based on these predictors yielded a larger area under the curve of 0.76 (95% CI: 0.70-0.82), but not significantly different from the BMRC (0.72: 95% CI: 0.65-0.77). CONCLUSIONS: This modern score is easy to apply and could be a sound predictor of poor prognosis. However, the availability of modern tests did not improve the ability to predict a bad outcome.

Weighing harm in therapeutic decisions of smear-negative pulmonary tuberculosis.

PURPOSE: To relate the intuitive weight of harm by commission and harm by omission in therapeutic decisions for pulmonary tuberculosis, and to compare it with a weight based on probabilities. METHODS: Clinicians were asked for an estimation of probabilities related with the outcome of treated and nontreated pulmonary tuberculosis and for the toll of wrong decisions. Three ratios of the weight of forgoing a treatment in false-negative patients against the weight of treating false-positives were calculated. The first was based on intuitive estimations, whereas the second and third were based on calculated, either through intuitive estimations of probabilities or through literature data. The association between experience and the difference between the intuitive and the calculated ratios was assessed. RESULTS: Eighty-one participants from Ecuador, Laos, Nepal, and Rwanda responded. The ratio of intuitive weights was 2.0 (interquartile range [IQR], 1.0-4.0) and the ratio of calculated weights based on intuitive probabilities was 64 (IQR, 25.0-169.6; P < 0.001). The ratio of calculated weight based on literature probabilities was 30 (IQR, 17.9-59.2). No association (R(2) = 0.03) was found between experience and accuracy in estimating the weight of errors. CONCLUSION: The weight of a false negative is more important than the weight of a false positive for therapeutic decisions in pulmonary tuberculosis. The ratio of the intuitively estimated weights was much lower than the calculation based on intuitively estimated influencing factors. Clinicians were accurate in estimating probabilities but failed to incorporate them into therapeutic decisions.

Bayesian clinical reasoning: does intuitive estimation of likelihood ratios on an ordinal scale outperform estimation of sensitivities and specificities?

RATIONALE: Bedside use of Bayes' theorem for estimating probabilities of diseases is cumbersome. An alternative approach based on five categories of powers of tests from 'useless' to 'very strong' has been proposed. The performance of clinicians using it was assessed. METHODS: Fifty clinicians attending a course of tropical medicine estimated powers of tests and post-test probabilities using the classical vs. the categorical Bayesian approach. The estimation of post-test probability was assessed for real and dummy diseases in order to avoid the bias of previous knowledge. Accuracy of answers was measured by the difference with reference values obtained from an expert system (Kabisa). RESULTS: Clinicians estimated positive likelihood ratios (LRs) a median of -1.07 log(10) lower than Kabisa [interquartile range (IQR): -1.47; -0.80] when derived classically and -0.17 (IQR: -0.42; +0.04) when estimated categorically (P < 0.001). For negative LRs the median was +0.39 log(10) higher (IQR: +0.71; +0.08) when derived classically and -0.18 log(10) lower (IQR: +0.03; -0.36) when estimated categorically (P < 0.001). Twenty (40%) disclosed not being able to calculate post-test probabilities using sensitivities and specificities. Regardless the approach post-test probabilities were overestimated both for real and dummy diseases [respectively +1.23 log(10) (IQR: +0,67; +2.08) and +2.03 log(10) (IQR: +0.49; +2.42)] (P = 0277), but the range was wider for the latter (P = 0.001). CONCLUSIONS: Participants were more accurate in estimating powers with a categorical approach than with sensitivities and specificities. Post-test probabilities were overestimated with both approaches. Knowledge of the disease did not influence the estimation of post-test probabilities. A categorical approach might be an interesting instructional tool, but the effect of training with this approach needs assessment.

Tuberculous meningitis: does lowering the treatment threshold result in many more treated patients?

OBJECTIVE: To determine how many more patients would be treated when lowering the treatment threshold for tuberculous meningitis. METHODS: From 1989 to 2004 findings of patients with symptoms lasting more than 1 week and inflammatory changes of cerebrospinal fluid (CSF) were collected. Several models of latent class analysis were tested. Cumulative numbers of cases were plotted against different cut-offs for post-test probability. RESULTS: In a cohort of 232 patients the prevalence of tuberculous meningitis (TBM) was estimated at 79.8% (95% CI. 67,0-88,1); probabilities above 80% were reached in 73% of patients. Lowering this threshold from 80% to 20% would add 14% more patients to be treated, for a total of 87%. A further lowering of the threshold to 5% would imply 5% more patients to be treated, bringing the cumulative number to 92%. The difference of lowering the threshold from 80% to 5% was 19%. CONCLUSION: In this setting, at least 75% of patients showing suggestive symptoms for more than a week and CSF changes very probably had TBM. The number of patients that should be treated does not increase linearly when lowering the threshold.