Respiratory Support Techniques for COVID-19-Related ARDS in a Sub-Saharan African Country: A Multicenter Observational Study.

BACKGROUND: Limited data from low-income countries report on respiratory support techniques in COVID-19-associated ARDS. RESEARCH QUESTION: Which respiratory support techniques are used in patients with COVID-19-associated ARDS in Uganda? STUDY DESIGN AND METHODS: A multicenter, prospective, observational study was conducted at 13 Ugandan hospitals during the pandemic and included adults with COVID-19-associated ARDS. Patient characteristics, clinical and laboratory data, initial and most advanced respiratory support techniques, and 28-day mortality were recorded. Standard tests, log-rank tests, and logistic regression analyses were used for statistical analyses. RESULTS: Four hundred ninety-nine patients with COVID-19-associated ARDS (mild, n = 137; moderate, n = 247; and severe, n = 115) were included (ICU admission, 38.9%). Standard oxygen therapy (SOX), high-flow nasal oxygen (HFNO), CPAP, noninvasive ventilation (NIV), and invasive mechanical ventilation (IMV) was used as the first-line (most advanced) respiratory support technique in 37.3% (35.3%), 10% (9.4%), 11.6% (4.8%), 23.4% (14.4%), and 17.6% (36.6%) of patients, respectively. The first-line respiratory support technique was escalated in 19.8% of patients. Twenty-eight-day mortality was 51.9% (mild ARDS, 13.1%; moderate ARDS, 62.3%; severe ARDS, 75.7%; P < .001) and was associated with respiratory support techniques as follows: SOX, 19.9%; HFNO, 31.9%; CPAP, 58.3%; NIV 61.1%; and IMV, 83.9% (P < .001). Proning was used in 79 patients (15.8%; 59 of 79 awake) and was associated with lower mortality (40.5% vs 54%; P = .03). The oxygen saturation to Fio2 ratio (OR, 0.99; 95% CI, 0.98-0.99; P < .001) and respiratory rate (OR, 1.07; 95% CI, 1.03-1.12; P = .002) at admission and NIV (OR, 6.31; 95% CI, 2.29-17.37; P < .001) or IMV (OR, 8.08; 95% CI, 3.52-18.57; P < .001) use were independent risk factors for death. INTERPRETATION: SOX, HFNO, CPAP, NIV, and IMV were used as respiratory support techniques in patients with COVID-19-associated ARDS in Uganda. Although these data are observational, they suggest that the use of SOX and HFNO therapy as well as awake proning are associated with a lower mortality resulting from COVID-19-associated ARDS in a resource-limited setting.

World Health Organization Surgical Safety Checklist: Compliance and Associated Surgical Outcomes in Uganda's Referral Hospitals.

BACKGROUND: A pilot study on the World Health Organization (WHO) Surgical Safety Checklist (SSC) showed a reduction in both major complications and mortality of surgical patients. Compliance with this checklist varies around the world. We aimed to determine the extent of compliance with the WHO SSC and its association with surgical outcomes in 5 of Uganda's referral hospitals. METHODS: A multicentre prospective cohort study was conducted in 5 referral hospitals in Uganda. Using a questionnaire based on the WHO SSC, patients undergoing surgical operations were systematically recruited into the study from April 2016 to July 2016. The patients were followed up daily for 30 days or until discharge for the purpose of documentation of complications. Logistic regression and linear regression were used to assess for association between compliance and perioperative surgical outcomes. RESULTS: We recruited 859 patients into the study. Overall compliance with the WHO SSC was 41.7% (95% confidence interval [CI], 39.7-43.8) ranging from 11.9% to 89.8% across the different hospitals. Overall compliance with "sign in" was 44.7% (95% CI, 43-45.6), with "time out" was 42.0% (95% CI, 39.4-44.6), and with "sign out" was 33.3% (95% CI, 30.7-35.9). There was no association between compliance and perioperative surgical outcomes: length of hospital stay, adverse events, and mortality. CONCLUSIONS: This study revealed low levels of compliance with the WHO SSC. There was a statistically significant association between this level of compliance and the incidence of pain and loss of consciousness postoperatively.

Analgesic Effects of Preincision Ketamine on Postspinal Caesarean Delivery in Uganda's Tertiary Hospital: A Randomized Clinical Trial.

Background. Good postoperative analgesic management improves maternal satisfaction and care of the neonate. Postoperative pain management is a challenge in Mulago Hospital, yet ketamine is accessible and has proven benefit. We determined ketamine's postoperative analgesic effects. Materials and Methods. We did an RCT among consenting parturients that were randomized to receive either intravenous ketamine (0.25 mg/kg) or placebo after spinal anesthetic. Pain was assessed every 30 mins up to 24 hours postoperatively using the numerical rating scale. The first complaint of pain requiring treatment was noted as "time to first breakthrough pain." Results. We screened 100 patients and recruited 88 that were randomized into two arms of 44 patients that received either ketamine or placebo. Ketamine group had 30-minute longer time to first breakthrough pain and lower 24-hour pain scores. Postoperative diclofenac consumption was lesser in the ketamine group compared to placebo and Kaplan-Meier graphs showed a higher probability of experiencing breakthrough pain earlier in the placebo group. Conclusion. Preincision intravenous ketamine (0.25 mg/kg) offered 30-minute prolongation to postoperative analgesia requirement with reduced 24-hour pain scores. We recommend larger studies to explore this benefit. This trial is registered with Pan African Clinical Trial Registry number PACTR201404000807178.

Feasibility of the modified sequential organ function assessment score in a resource-constrained setting: a prospective observational study.

BACKGROUND: Sub-Saharan Africa has a great burden of critical illness with limited health care resources. We evaluated the feasibility and utility of the modified Sequential Organ Function Assessment (mSOFA) score in assessing morbidity and mortality in the National Referral Hospital's intensive care unit (ICU) for one year. METHODS: We conducted a prospective, observational cohort study on patients above 12 years of age admitted to the ICU at Mulago Hospital (Kampala, Uganda). All SOFA scores were determined at admission and at 48 h. We modified the SOFA score by replacing the PaO2/FiO2 ratio with SPO2/FiO2. The primary outcome was ICU mortality. RESULTS: This ICU cohort of 118 patients had a mean age of 37 years and an ICU mortality rate of 46.6%. Non-survivors had higher initial (7.7 SD 3.8 vs. 5.5 SD 3.3; p = 0.007), mean (8.1 SD 3.9 vs 4.7 SD 2.6; p < 0.001) and highest mSOFA scores (9.4 SD 4.2 vs. 5.8 SD 3.2; p < 0.001), with an increase of 1.0 (SD 3.1) mSOFA on average after 48 h when compared to survivors (p < 0.001). The area under the receiver operating characteristic curves for each mSOFA category was: initial-0.68, mean-0.76, highest-0.76 and delta mSOFA-0.74. Multivariate logistic regression analysis showed no significant association between mSOFA scores and mortality. CONCLUSION: Our results confirm that calculation of the mSOFA score is feasible for an ICU population in a resource-limited country. More data are needed to test for an association between mSOFA and mortality.

Incidence and Risk Factors for Delirium among Mechanically Ventilated Patients in an African Intensive Care Setting: An Observational Multicenter Study.

Aim. Delirium is common among mechanically ventilated patients in the intensive care unit (ICU). There are little data regarding delirium among mechanically ventilated patients in Africa. We sought to determine the burden of delirium and associated factors in Uganda. Methods. We conducted a multicenter prospective study among mechanically ventilated patients in Uganda. Eligible patients were screened daily for delirium using the confusional assessment method (CAM-ICU). Comparisons were made using t-test, chi-squares, and Fisher's exact test. Predictors were assessed using logistic regression. The level of statistical significance was set at P < 0.05. Results. Of 160 patients, 81 (51%) had delirium. Median time to onset of delirium was 3.7 days. At bivariate analysis, history of mental illness, sedation, multiorgan dysfunction, neurosurgery, tachypnea, low mean arterial pressure, oliguria, fevers, metabolic acidosis, respiratory acidosis, anaemia, physical restraints, marital status, and endotracheal tube use were significant predictors. At multivariable analysis, having a history of mental illness, sedation, respiratory acidosis, higher PEEP, endotracheal tubes, and anaemia predicted delirium. Conclusion. The prevalence of delirium in a young African population is lower than expected considering the high mortality. A history of mental illness, anaemia, sedation, endotracheal tube use, and respiratory acidosis were factors associated with delirium.