School of Medicine (Sch. of Med.) Collections

Permanent URI for this collection

https://makir.mak.ac.ug/handle/10570/183

Browse

Recent Submissions

Now showing 1 - 5 of 1340
  • Item
    Incidence and risk factors for admissions among children attending the Sickle Cell Clinic at Mulago National Referral Hospital: A prospective cohort study nested in the SPARCO registry
    (Makerere University, 2026) Ssegawa, Lawrence
    Background: Despite improvements in care, sickle cell disease (SCD) remains a major cause of childhood morbidity and mortality in sub-Saharan Africa, with high rates of hospitalization. This study aimed to determine the incidence of hospitalizations and identify associated risk factors among children attending the Mulago National Referral Hospital Sickle Cell Clinic in the hydroxyurea era. Methods: This prospective observational cohort study was nested within the Sickle PanAfrican Research Consortium (SPARCO) Uganda Registry. Children aged 6 months to 18 years enrolled in the registry between June 2025 and January 2026 were enrolled and followed for six months. Data on socio-demographic characteristics, clinical history, maintenance treatment (including hydroxyurea adherence), and prior hospital admissions collected at baseline, supplemented by data from the registry. Enrolled children were followed up for 6 months and assessed for ill health requiring hospitalization. Incidence density was calculated as admissions per 1,000 participant-years, while time to first admission was analysed using Kaplan-Meier survival curves. Risk factors were evaluated using multivariable Cox proportional hazards regression, with adjusted hazard ratios (aHR) and 95% confidence intervals (CI) reported. Statistical significance was set at p < 0.05. Results: A total of 405 children and the median age was 5.0 years (IQR 3.0-8.0), and 205 were male. Of the cohort, 316 (78.0%) were on hydroxyurea at baseline, however, only 57% had good adherence. Overall, a total of 142 occurred translating into an incidence density of 701 admissions per 1,000 participant-years (95% CI 592–828). A total of 142 (35.1%) had at least one hospitalization while 30 (21.1%) had two admissions during the study period. Vasoocclusive pain crises (69.0%: 98/142), severe anaemia (29.6%: 42/142), and acute chest syndrome (21.8%: 31/142) were the most common reasons for hospitalization. The probability of remaining admission-free was 78.2% at three months and 64.9% at six months. In multivariable analysis, younger age (<5 years) (aHR 1.78, 95% CI 1.24–2.56, p=0.002), lower caretaker education (aHR 1.45, 95% CI 1.10–1.92, p=0.009), greater distance to facility (>10 km) (aHR 1.62, 95% CI 1.18–2.23, p=0.003), and history of frequent crises (aHR 2.10, 95% CI 1.45–3.04, p<0.001) independently increased the risk of admission. Hydroxyurea use (aHR 0.52, 95% CI 0.34–0.79, p=0.002) and high baseline haemoglobin (aHR 0.78 per g/dL, 95% CI 0.68–0.90, p<0.001) were strongly protective. inadequate knowledge, persistent pain, poor communication, and drug stock-outs. Overall, 28 children (6.9%) were lost to follow-up and 2 died (both from acute chest syndrome). Conclusion: Despite widespread hydroxyurea utilisation, hospitalisation rates remain high among Ugandan children with SCD, driven primarily by vaso-occlusive crises and infections. Younger age and socioeconomic/access barriers increase risk, while hydroxyurea and better haemoglobin status are protective. Keywords: Sickle cell disease, hospitalisation, incidence, risk factors, hydroxyurea, Uganda, SPARCO registry
  • Item
    Patterns of cardiac diseases and risk factors forin-hospital mortality among children admitted to Mulago National Referral Hospital
    (Makerere University, 2026) Akol, Christine
    Background: Paediatric cardiac diseases, including congenital and acquired heart conditions, are a major cause of morbidity and mortality in low-resource settings such as Uganda. Despite the dual burden of congenital heart disease (CHD) and preventable acquired conditions like rheumatic heart disease, local evidence on disease patterns and determinants of in-hospital mortality remains limited. Objective: To describe the patterns of cardiac diseases, estimate the in-hospital mortality, and identify risk factors for mortality among children admitted to Mulago National Referral Hospital. Methods: A retrospective cohort study was conducted using medical records of children aged 0–17 years admitted to the cardiopulmonary ward (Firm C) with confirmed cardiac disease from January 2015 to January 2025. Demographic, clinical, laboratory, and echocardiographic data were extracted. Disease patterns were summarized descriptively. In-hospital mortality was expressed as a proportion and incidence density with 95% confidence intervals (CI). Cox proportional hazards regression and Kaplan-Meier curves were used for survival analysis. Results: A total of 521 children were included, with a mean age was 3.0 ± 4.2 years; 41.8% were infants aged 1-12 months and 50.3% were male. Congenital heart disease accounted for 81.6% of admissions, predominantly ventricular septal defect (32.2%), tetralogy of Fallot (17.1%), and atrial septal defect (11.9%). Acquired heart disease comprised 18.4% of admissions, mainly rheumatic heart disease (8.8%) and dilated cardiomyopathy (5.4%). The overall in-hospital mortality was 21.1% (110/521; 95% CI: 17.6–24.9). The incidence density of mortality was 25.2 deaths per 1000 persondays (95% CI 20.7-30.4), with median hospital stay of 6.0 days (IQR: 3.0–11.0). Independent predictors of mortality were malnutrition (aHR 2.38, 95% CI: 1.28-4.42), heart failure (aHR 2.61, 95% CI: 1.41-4.83), severe pneumonia (aHR 2.89, 95% CI: 1.48-5.64), and lower platelet count (aHR 0.75 per 100×10⁹/L increase, 95% CI: 0.60-0.94). Cardiac diagnosis was not independently associated with mortality. Conclusion: Children admitted with cardiac disease at Mulago National Referral Hospital experience substantial in-hospital mortality. Mortality was primarily associated with potentially modifiable clinical conditions including malnutrition, heart failure and severe pneumonia rather than the underlying cardiac diagnosis itself. Strengthening early recognition and integrated management of these comorbidities offers an immediate opportunity to improve inpatient survival even within current resource constraints. Keywords: congenital heart disease, acquired heart disease, rheumatic heart disease, paediatric cardiology, mortality, heart failure, Uganda.
  • Item
    Prevalence and factors associated with limb length discrepancy following intramedullary nailing of femur shaft fractures in adults at Mulago National Referral Hospital
    (Makerere University, 2026) Kasirye, Ronald
    Background. Limb length discrepancy (LLD) is a recognized complication following intramedullary nailing (IMN) of femoral shaft fractures, with a reported global incidence of 20–43%. In sub-Saharan Africa, limited access to intraoperative imaging, and fracture tables heightens this risk. In Uganda, the burden of LLD following IMN remains poorly characterized, creating an evidence gap. Complications such as gait disturbance, early arthritis and psychological distress may occur in patients with LLD. Objectives. The objective of this study was to determine the prevalence of LLD and to identify the factors associated with LLD following IMN of femoral shaft fractures among adults at Mulago National Referral Hospital (MNRH), Kampala, Uganda. Methods. An analytical cross-sectional study was conducted at Mulago National Referral Hospital from October 2025 to February 2026. 101 consecutive adult patients (≥18 years) who had undergone IMN for femur shaft fractures were enrolled on their first post-operative day. Limb length was measured radiologically using CT-scanogram. LLD was categorized as mild LLD (<2 cm), moderate (2–5 cm), or severe (>5 cm). Ordinal logistic regression adjusted for clustering at the surgical team level was used for bivariate and multivariable analysis in STATA version 15. Results. The overall prevalence of LLD was 38.6% [95% CI (29.1,49.8)], 23.8%[95% CI (15.9,33.3)] had mild LLD and 14.8% [95% CI (8.6,23.3)] had moderate LLD; In multivariable ordinal logistic regression, greater trochanter nail entry (aOR 2.79, 95% CI 1.99,3.90, p<0.001), absence of pre-operative traction (aOR 3.96, 95% CI 3.04,5.17, p<0.001), comminuted fractures with (aOR 13.62, 95% CI 2.14,86.74, p=0.006), Fracture location (mid-third aOR 0.21; distal aOR 0.12, both p<0.001), Mid-thigh circumference (aOR 1.11, 95% CI 1.07,1.14, p<0.001),were independently associated with LLD. Surgical delay was strongly associated with worse LLD in bivariate analysis (OR 84.00, 95% CI 21.73,324.64, p<0.001) though model instability due to sparse data. precluded its inclusion in the final multivariable model. Conclusions. LLD is a complication following femur fractures after IMN at MNRH, affecting 38.6% of patients on the first post-operative day. Complex fracture patterns, absence of pre-operative traction, and greater trochanter nail entry were independently associated with LLD. Reducing time to surgery, standardizing pre-operative traction for all patients awaiting IMN, and improving access to intraoperative fluoroscopy are the most actionable interventions to reduce the LLD burden at this institution.
  • Item
    Skeletal-related events in multiple myeloma at Uganda Cancer Institute: prevalence, patterns, associated factors and overall survival
    (Makerere University, 2026) Baguma, Ezekiel
    Background: Skeletal-related events are a major source of morbidity in multiple myeloma, yet their burden, anatomical distribution, and prognostic significance remain poorly characterized in sub-Saharan Africa, where patients typically present with advanced disease. Uganda Cancer Institute is the sole national oncology referral facility in Uganda, and its growing multiple myeloma caseload from 65 patients in 2021 to 117 in 2024 highlights an urgent need for locally derived evidence to guide clinical management. Objective: To determine the prevalence, patterns, and factors associated with skeletal-related events among multiple myeloma patients at Uganda Cancer Institute and to assess their two-year overall survival. Methods: A retrospective cohort study was conducted on 181 adults with confirmed multiple myeloma diagnosed at Uganda Cancer Institute between January 2021 and December 2023. Independent factors associated with skeletal-related event occurrence were identified using multivariable modified Poisson regression, reported as adjusted prevalence ratios (aPR) with 95% confidence intervals. Two-year overall survival was estimated using the Kaplan-Meier method, and independent predictors of mortality identified using Cox proportional hazards regression. Results: Skeletal-related events occurred in 71.8% of patients (n=130; 95% CI 65.2–78.4), with 60.8% already present at admission. Pathological fractures (60.8%) and spinal cord compression (37.6%) were the most frequent skeletal-related event types, both predominantly involving the thoracic and lumbar spine. Durie-Salmon Stage III disease (aPR 1.73, 95% CI 1.19–2.51, p=0.004) and HIV co-infection (aPR 1.32, 95% CI 1.10–1.58, p=0.003) were independently associated with skeletal-related event occurrence. Two-year survival was significantly lower in patients with skeletal-related events than without (33% vs 55%; log-rank p=0.022). On multivariable Cox regression, the skeletal-related event mortality association was attenuated after adjustment (aHR 1.45, p=0.276). Conclusion: Skeletal-related events are highly prevalent at Uganda Cancer Institute, affecting 71.8% of patients, driven by late-stage presentation, with advanced disease stage and HIV co-infection as independent associated factors. These findings highlight the need for earlier multiple myeloma diagnosis at lower health system levels, routine spinal imaging for patients presenting with back pain or neurological symptoms, and a national multidisciplinary guideline for multiple myeloma management in Uganda.
  • Item
    Prevalence of neglected long bone fractures and associated factors among patients with musculoskeletal trauma at Mulago National Referral Hospital
    (Makerere University, 2026) Dembe, Moses
    Background: Neglected long bone fractures, defined as fractures presenting more than four weeks after injury without prior definitive orthopaedic management, are a major cause of preventable disability in low- and middle-income countries. Their burden in Uganda is poorly quantified, and the patient and health-system factors driving prolonged neglect at the Mulago National Referral Hospital (MNRH) have not been systematically described. Objectives: To determine the prevalence of neglected long bone fractures among patients presenting with long bone fractures at Mulago National Referral Hospital (MNRH), and to identify factors associated with prolonged neglect among those patients. Methods: A hospital-based, cross-sectional analytical study with a nested screening component was conducted at the orthopaedic outpatient clinic and accident and emergency unit of MNRH. All long bone fracture patients presenting during the study period were screened and logged to provide the prevalence denominator. Patients meeting the criteria for neglect were enrolled consecutively until the target sample size of 105 was reached. Data were collected using a structured, pre-tested, bilingual (English/Luganda) questionnaire and a focused clinical and radiological assessment. The period prevalence was reported with a 95% confidence interval. Factors associated with prolonged neglect (≥90 days from injury to presentation) among neglected patients were examined using bivariate logistic regression for crude associations and Firth’s penalized logistic regression for the multivariable model, with the final model built by backward elimination. Results: During the study period, 843 patients with long bone fractures were evaluated, of whom105 had neglected fractures, giving a prevalence of 12.5%. The median duration of neglect was 62 days (IQR 38–150), with the majority (70.5%) delaying care for 1–3 months. Young adults experienced longer delays, with 38.9% presenting after 91–180 days, while children and the elderly presented earlier (30–90 days). Despite delays, all participants sought some form of care within the first 28 days, predominantly from traditional bone setters (73.3%), whereas only 26.7% visited local clinics. The tibia/fibula was the most commonly fractured bone (38.1%), with fractures occurring more frequently on the left side (63.8%), mainly in the distal segment (39.0%), and predominantly closed (90.5%). Multivariate analysis identified several factors independently associated with neglected fractures: male sex (AOR = 31.7, P=0.001), alcohol use (AOR = 11.8, P=0.007), residence >10 km from a health facility (AOR = 5.7, P=0.013), lack of prior medical advice (AOR = 6.4, P=0.007), and difficulty accessing orthopaedic care (AOR = 13.2, P=0.001). Conclusion: Neglected long bone fractures are common at MNRH and disproportionately affect young adult men. Prolonged neglect is driven jointly by behavioural factors and by structural and informational barriers within the health system, and universal early contact with traditional bone setters represents the single most promising point of intervention. Targeted public-health messaging, structured engagement with traditional bone setters, expansion of regional orthopaedic capacity, and a dedicated rapid-triage pathway for late-presenting patients at MNRH are urgently needed.