HIV-positive patients in the intensive care unit: A retrospective audit.

BACKGROUND: The indications for and outcomes of intensive care unit (ICU) admission of HIV-positive patients in resource-poor settings such as sub-Saharan Africa are unknown. OBJECTIVE: To identify indications for ICU admission and determine factors associated with high ICU and hospital mortality in HIV-positive patients. METHODS: We reviewed case records of HIV-positive patients admitted to the medical and surgical ICUs at Groote Schuur Hospital, Cape Town, South Africa, from 1 January 2012 to 31 December 2012. RESULTS: Seventy-seven HIV-positive patients were admitted to an ICU, of whom two were aged <18 years and were excluded from the final analysis. HIV infection was newly diagnosed in 37.3% of the patients admitted during the study period. HIV-positive patients had a median CD4 count of 232.5 (interquartile range 59 - 459) cells/µL. Respiratory illness, mainly community-acquired pneumonia, accounted for 30.7% of ICU admissions. ICU and hospital mortality rates were 25.3% and 34.7%, respectively. Predictors of ICU mortality included an Acute Physiology and Chronic Health Evaluation ΙΙ (APACHE II) score >13 (odds ratio (OR) 1.4, 95% confidence interval (CI) 1.1 - 1.7; p=0.015), receipt of renal replacement therapy (RRT) (OR 2.2, 95% CI 1.2 - 4.1; p=0.018) and receipt of inotropes (OR 2.3, 95% CI 1.6 - 3.4; p<0.001). Predictors of hospital mortality were severe sepsis on admission (OR 2.8, 95% CI 0.9 - 9.1; p=0.07), receipt of RRT (OR 1.9, 95% CI 1.0 - 3.6; p=0.056) and receipt of inotropic support (OR 2.0, 95% CI 1.4 - 3.2; p<0.001). Use of highly active antiretroviral therapy (HAART), CD4 count, detectable HIV viral load and diagnosis at ICU admission did not predict ICU or hospital mortality. CONCLUSIONS: Respiratory illnesses remain the main indication for ICU in HIV-positive patients. HIV infection is often diagnosed late, with patients presenting with life-threatening illnesses. Severity of illness as indicated by a high APACHE ΙΙ score, multiple organ dysfunction requiring inotropic support and RRT, rather than receipt of HAART, CD4 count and diagnosis at ICU admission, are predictors of ICU and hospital mortality.

HIV-positive patients in the intensive care unit: A retrospective audit

Background. The indications for and outcomes of intensive care unit (ICU) admission of HIV-positive patients in resource-poor settings such as sub-Saharan Africa are unknown.Objective. To identify indications for ICU admission and determine factors associated with high ICU and hospital mortality in HIV-positive patients.Methods. We reviewed case records of HIV-positive patients admitted to the medical and surgical ICUs at Groote Schuur Hospital, Cape Town, South Africa, from 1 January 2012 to 31 December 2012.Results. Seventy-seven HIV-positive patients were admitted to an ICU, of whom two were aged <18 years and were excluded from the final analysis. HIV infection was newly diagnosed in 37.3% of the patients admitted during the study period. HIV-positive patients had a median CD4 count of 232.5 (interquartile range 59 - 459) cells/µL. Respiratory illness, mainly community-acquired pneumonia, accounted for 30.7% of ICU admissions. ICU and hospital mortality rates were 25.3% and 34.7%, respectively. Predictors of ICU mortality included an Acute Physiology and Chronic Health Evaluation ΙΙ (APACHE II) score >13 (odds ratio (OR) 1.4, 95% confidence interval (CI) 1.1 - 1.7; p=0.015), receipt of renal replacement therapy (RRT) (OR 2.2, 95% CI 1.2 - 4.1; p=0.018) and receipt of inotropes (OR 2.3, 95% CI 1.6 - 3.4; p<0.001). Predictors of hospital mortality were severe sepsis on admission (OR 2.8, 95% CI 0.9 - 9.1; p=0.07), receipt of RRT (OR 1.9, 95% CI 1.0 - 3.6; p=0.056) and receipt of inotropic support (OR 2.0, 95% CI 1.4 - 3.2; p<0.001). Use of highly active antiretroviral therapy (HAART), CD4 count, detectable HIV viral load and diagnosis at ICU admission did not predict ICU or hospital mortality.Conclusions. Respiratory illnesses remain the main indication for ICU in HIV-positive patients. HIV infection is often diagnosed late, with patients presenting with life-threatening illnesses. Severity of illness as indicated by a high APACHE ΙΙ score, multiple organ dysfunction requiring inotropic support and RRT, rather than receipt of HAART, CD4 count and diagnosis at ICU admission, are predictors of ICU and hospital mortality

Determinants of dynamic hyperinflation during metronome-paced tachypnea in COPD and normal subjects.

In COPD, dynamic hyperinflation (DH) occurs during exercise and during metronome-paced tachypnea (MPT). We investigated the relationship of DH with breathing pattern and ventilation (V˙E) in COPD and normal subjects (NS). In 35 subjects with moderate COPD and 17 younger healthy volunteers we measured inspiratory capacity (IC), breathing frequency (fR), expiratory time (TE), ventilation (V˙E) and end-tidal carbon dioxide tension (PETCO2) at baseline and after 30s of MPT at 40breaths/min with metronome-defined I:E ratios of 1:1 and 1:2. A reduction in IC (ΔIC) was taken to indicate DH. In COPD subjects, DH correlated with TE but not with V˙E or PETCO2, and was best predicted by total lung capacity. NS also showed DH (although less than in COPD), which correlated with PETCO2 but not with fR, TE or V˙E. We conclude that MPT evokes DH in both NS and patients with COPD. TE is the most important determinant of DH during MPT in patients with COPD.

Asthma and hyperthyroidism. A report of 4 cases.

For many years it has been known that thyrotoxicosis can worsen asthma, increasing both the frequency and severity of asthma attacks and increasing requirements for medication. Clinical recognition of this association may be difficult. Thyrotoxic asthmatics frequently experience side-effects from medications, particularly beta-receptor agonists and theophyllines. Four case reports of asthmatics in whom thyrotoxicosis was diagnosed are presented. All 4 patients improved on returning to euthyroid status. The mechanism responsible for the harmful interaction is not known.

Transient oliguria with renal tubular dysfunction after a 90 km running race.

In the course of a 19-d study of renal function in five ultramarathon runners, before, during and after a 90 km race, one runner developed transient oliguria with renal tubular dysfunction and anuria during and immediately after the race. Other features of the renal failure were an 84-fold increase in urine beta 2-microglobulin excretion (from 0.19 to 16.0 micrograms.min-1) and a much smaller increase in urine total protein excretion (from 0.07 to 0.18 mg.min-1) during the post-race period. Post-race creatinine clearance remained below pre-race levels throughout the study, varying between 42.8 and 72.9 ml.min-1, in contrast to the post-race 49% increase in the remaining runners (from 138.1 +/- 12.9 to 205.5 +/- 59.9 ml.min-1). Osmolal clearance also remained low (0.31 to 0.98 ml.min-1) compared with the pre-race values (1.46 +/- 0.02 ml.min-1), as did the urine flow rates (0.11 to 0.18 ml.min-1) compared with the pre-race values (0.34 +/- 0.02 ml.min-1). This renal dysfunction persisted despite the patient receiving 2 l of intravenous fluids immediately after the race and probably resulted from fluid restriction during the race. There was full recovery of renal function 1 yr later when the subject again ran the Comrades Marathon.

Muscle glycogen repletion during active postexercise recovery.

High-intensity intermittent bicycle exercise was used to deplete muscle glycogen levels by 70% and elevate blood lactate levels to greater than 13.0 mmol/l. Thereafter subjects either cycled with one leg for 45 min followed by 45 min of passive recovery (partially active recovery) or rested for 90 min (passive recovery). During the first 45 min of partially active recovery 1) blood lactate (P less than 0.05) and pH levels (P less than 0.05) returned more rapidly to preexercise values than during passive recovery, 2) the rate of net glycogen resynthesis (0.28 mumol . g-1 . min-1) was the same in both legs, and 3) muscle lactate levels were significantly lower (P less than 0.05) in the passive than in the active leg. Thereafter the rate of net muscle glycogen resynthesis was unchanged (0.26 mumol . g-1 . min-1) and lactate removal could theoretically account for only 18% of the glycogen resynthesized. Overall, the rate of muscle glycogen resynthesis and muscle lactate removal was not different from that measured during passive recovery. After high-intensity exercise 1) glycogen repletion is not impeded by light exercise, and 2) blood glucose is an important substrate for glycogen resynthesis.