Introduction
Cardiovascular diseases are the leading cause of global mortality. Reference Mensah, Fuster and Murray1 Inpatient cardiac care frequently involves antibiotic exposure for peri-procedural prophylaxis and suspected infection. Monitoring antibiotic use is essential for stewardship, particularly in high-acuity low- and middle-income country settings, where empirical practice may accelerate resistance. Reference Shomuyiwa, Lucero-Prisno and Manirambona2 Tanzania has national antibiotic consumption estimates from regulatory data sets. Reference Sangeda, Saburi and Masatu3,Reference Sangeda, William and Masatu4 However, facility-level utilization in specialized cardiovascular settings remains sparse. The Jakaya Kikwete Cardiac Institute (JKCI) is Tanzania’s only dedicated cardiac referral center equipped with catheterization and open-heart surgical capabilities. We analyzed six fiscal years of inpatient dispensing records to describe antibiotic use patterns, the distribution of antibiotic classes, and the World Health Organization (WHO) Access–Watch–Reserve (AWaRe) categorization among cardiovascular inpatients. Reference Mutatina, Lomnyak and Bizimana5
Materials and methods
This retrospective longitudinal analysis used inpatient systemic antibiotic dispensing records from July 2016 to June 2022 at the JKCI (150 beds) in Dar es Salaam, Tanzania. Routine dispensing data were extracted from the MedPro system. Only systemic antibiotics (oral and parenteral) were included in this study.
Antibiotics were classified by anatomical therapeutic chemical (ATC) level 5 and by the 2021 WHO-AWaRe categories. Analyses were descriptive and inferential (SPSS v26); group differences in mean defined daily doses (DDD) per 100 bed-days were assessed using one-way ANOVA. There were no facility-specific antibiotic guidelines during the study period, and Tanzania’s National Antimicrobial Resistance National Action Plan was launched in 2017. Reference Mutatina, Lomnyak and Bizimana5 Thus, no stewardship framework was yet operationalized at JKCI.
Antibiotic use was measured in DDD per 100 bed-days 6 standardized to annual occupancy. An occupancy-corrected denominator was used to account for the bed count and fraction of each fiscal year in active clinical service using the formula :
$$ {\rm{Total\; DDD\;\times100}} \over{\rm{days\;admitted\;\times\;days\;in\;year\;serviced\;\times\;total\;beds}} $$
This adjustment was necessary because the JKCI was scaling up toward full capacity during its early stages of development. The fiscal years 2016–2017 had only 182.5 serviced days and 63 inpatients receiving antibiotics, explaining the lower numeric DDD values in the first year.
Expanded tables are available in the associated public preprint (medRxiv doi:10.1101/2025.09.24.25336595).
Ethical approval was obtained from the Muhimbili University of Health and Allied Sciences Research Ethics Committee and from JKCI. All data were anonymized and contained no identifiable patient information.
Results
A total of 30,885 inpatient antibiotic prescriptions from 6,612 cardiovascular inpatients were analyzed. Overall use was 29.88 DDD per 100 bed-days (Table 1). Carbapenems were the dominant ATC class and meropenem was the leading molecule (Figure 1). Reserve agents comprised 47.0% of all use (14.04 DDD/100 bed-days), Watch 31.7% (9.46) and Access 21.3% (6.38). Injectable formulations dominated (16.57 DDD per 100 bed-days; 55.5%), followed by tablets (9.59; 32.1%), syrups (2.06; 6.9%) and capsules (1.66; 5.6%).
Figure 1. Top 12 Anatomical therapeutic chemical (ATC) level 5 antibiotic molecules contributing 91.4% of total inpatient use, expressed as defined daily doses (DDD) per 100 bed-days at Jakaya Kikwete Cardiac Institute, Tanzania, in 2016–2022.
Table 1. Patient characteristics and inpatients’ antibiotic use indicators at the Jakaya Kikwete Cardiac Institute, Tanzania, in 2016–2022. P-values indicate differences in mean defined daily doses (DDD) per 100 bed-days across categories within each variable, using ANOVA
Discussion
Antibiotic use among cardiac inpatients was high and dominated by broad-spectrum and last-line agents. Carbapenems and third-generation cephalosporins accounted for most of the DDD volume. Similar patterns have been reported in African tertiary hospitals. Reference Shomuyiwa, Lucero-Prisno and Manirambona2,Reference Mutatina, Lomnyak and Bizimana5 The heavy reliance on injectables underscores opportunities for IV-to-oral switch, and the AWaRe imbalance—with Reserve and Watch accounting for 78.7% of all use—signals a clear stewardship priority. 7 Empirical meropenem defaulting is probable in the absence of local guidelines and structured escalation pathways. Reference Zaroff, Cheetham and Palmetto8 Use of an occupancy-normalized denominator was essential for correctly interpreting the early partial year. 6 Contextually, these findings sit within global increases in antibiotic consumption Reference Van Boeckel, Gandra and Ashok9 and global AMR burden estimates. Reference Murray, Ikuta and Sharara10 JKCI therefore provides a benchmark for stewardship design in specialized cardiovascular inpatient care.
Conclusion
Cardiovascular inpatients at Tanzania’s national cardiac center received substantial antibiotic exposure, dominated by Reserve and Watch antibiotics. Strengthening stewardship through AWaRe-aligned formulary oversight and IV-to-oral protocols can reduce unnecessary broad-spectrum use.
Open access
