Discussion

Between 2017 and 2025, the prevalence of HAIs at Hedi Chaker University Hospital increased significantly, peaking in 2025. BSI, skin and soft tissue infections, and MDR pathogens also showed significant increases over time. Pneumonia and lower respiratory tract infections were the most frequent HAIs, although Gram-negative bacteria were the predominant pathogens identified. In addition, peripheral venous catheterization, undernutrition, and intensive care unit admission were independently associated with the occurrence of HAI.

The upward trajectory of HAI prevalence is consistent with data from Tunisian university hospitals, where repeated PPS showed HAI prevalence rising significantly from 12.3% in 2012 to 15.5% in 2020, ^{Reference Ghali, Ben Cheikh, Bhiri, Khefacha, Latiri and Ben Rejeb6} and with southern Tunisian hospitals reporting a prevalence of 8.4% in 2023. ^{Reference Baklouti, Ben Hmida and Ben Ayed7} The gap between high-income and low-to middle-income countries remains substantial, with African studies reporting pooled HAI prevalence ranging from 7.24% to 28%, more than twice European rates, largely attributable to structural deficits in IPC infrastructure and the absence of national surveillance systems. ^{Reference Bunduki, Masoamphambe, Fox, Musaya, Musicha and Feasey8,Reference Letaief, Rejaibi and Hechaichi9} International comparisons suggest that HAI trends vary across healthcare settings, with differences in IPC resources potentially contributing to this heterogeneity. In Europe, adjusted HAI prevalence increased modestly from 6.5% in 2016–2017 to 8.0% in 2022–2023. ² In contrast, a previous national US survey reported a decline in HAI prevalence from 3.2% in 2015 to 2.6% in 2023. ^{Reference Chea, Li and Eure19} Similarly, a four-year repeated PPS conducted in an Italian tertiary hospital found no significant upward trend in HAI prevalence between 2022 and 2025. ^{Reference Pennisi, Godoy and Camuffo20} In our setting, the observed rising trajectory should therefore be interpreted in the context of comparatively limited IPC resources. The sharp rise observed in 2025 likely reflects both the enduring consequences of the COVID-19 pandemic, during which antimicrobial-resistant HAIs increased by up to 78% for some pathogens between 2019 and 2020, ¹⁰ and the progressive accumulation of antimicrobial resistance pressure within the hospital environment. The sudden surge of the 2025 peak suggests an acute determinant over gradual risk accumulation. MDR isolates rose from 28.6% in 2024 to 66.7% in 2025, a sharper shift than the overall HAI trend, consistent with a localized outbreak in one or more high-risk units where device use and broad-spectrum antibiotic exposure favor resistant-strain dissemination. The concurrent rise in bloodstream infections, largely caused by MDR Gram-negative bacilli in our setting, supports this interpretation. Another explanation for this sharp rise could be a concurrent supply chain breakdown that caused an acute shortage of hospital-grade surface and hand disinfectants during this period.

The pathogen profile observed in our cohort, mainly a predominance of E. coli, P. aeruginosa, and K. pneumoniae, broadly mirrors both African and European surveillance data, though with differing relative rankings. A recent African systematic review similarly reported E. coli (18.3%), S. aureus (17.3%), and Klebsiella spp. (17.2%) as the leading HAI pathogens, ^{Reference Bunduki, Masoamphambe, Fox, Musaya, Musicha and Feasey8} whereas the 2022–2023 ECDC PPS identified E. coli as the single most common European isolate (13%), followed by Klebsiella spp. and Enterococcus species. ² The relative overrepresentation of P. aeruginosa in our series may reflect a higher local burden of device-associated and ICU-acquired infections.

The significant increase in BSIs mirrors international surveillance data: in Canadian acute care hospitals, BSI prevalence rose from 1.0% in 2017 to 1.5% in 2024, with central line-associated BSIs doubling over the same period. ^{Reference Mitchell, Lee and Bartoszko11} In critically ill COVID-19 patients admitted to a Tunis medical ICU, BSIs constituted a major HAI subtype. ^{Reference Trifi, Sellaouti and Mehdi12} The rising BSI trend most plausibly reflects the increasing use of invasive vascular devices in complex patients, combined with an insufficient catheter care bundle implementation. Both phenomena are, however, deeply interconnected with the rising MDR burden: MDR Gram-negative bacilli—including ESBL-producing Enterobacterales and carbapenem-resistant strains—readily colonize invasive vascular devices and evade standard empirical regimens, creating a direct pathway to device-associated BSIs. This association is well documented in the Tunisian context, where MDR organisms have been consistently identified as the predominant pathogens in hospital-acquired bloodstream infections over the past decade. ^{Reference Trifi, Sellaouti and Mehdi12,Reference Kanzari, Ferjani and Meftah13} The postCOVID-19 period further amplified this dynamic, as the widespread use of broad-spectrum antibiotics in critically ill patients accelerated resistance selection, ^{Reference Horan, Andrus and Dudeck5} sustaining a cycle between MDR emergence and BSI incidence that the trends observed in our series clearly illustrate.

Part of the increase observed in the “Other” HAI category may be attributable to ectoparasitic infestations, namely scabies and pediculosis, although these were not recorded as a distinct infection site and their individual trend could not be quantified. Parasitic infections more broadly are known to play a considerable role in HAIs, particularly in an increasingly complex hospital environment with a growing proportion of immunocompromised patients. ^{Reference Fürnkranz and Walochnik14} The nonspecific cutaneous manifestations of scabies frequently lead to delayed diagnosis and sustained nosocomial transmission, ^{Reference Kim, Kim and Yun15} and overcrowded wards with close patient contact may have created favorable conditions for transmission in our setting.

The three independent risk factors identified in the multivariate analysis—peripheral venous catheter use, undernutrition, and ICU admission—are well documented in both Tunisian and international surveys. ICU-admitted patients carry the highest HAI risk, given the convergence of severe underlying illness, multiple invasive devices, and intensive antibiotic exposure; national multicenter data from Tunisian ICUs reported an HAI prevalence of 25.2%, further corroborating this gradient. ^{Reference Jamoussi, Ayed and Ben Ismail16} Undernutrition compromises mucosal integrity and immune defense mechanisms, rendering affected patients disproportionately vulnerable to nosocomial pathogens, a finding consistently reported across PPS conducted in comparable settings, and addressing it has been proposed as an integral component of HAI reduction strategies. ^{Reference Ghali, Ben Cheikh, Bhiri, Khefacha, Latiri and Ben Rejeb6,Reference Baklouti, Ben Hmida and Ben Ayed7,Reference Fitzpatrick, Skally and O’Hanlon17} The identification of peripheral venous catheter use as an independent predictor is noteworthy: a systematic review showed that peripheral venous catheters account for a mean of 23% of nosocomial catheter-related bloodstream infections, with prolonged dwell time and emergent insertion as the main risk factors—yet this device remains systematically under-surveilled compared to central venous access. ^{Reference Mermel18}

To our knowledge, this is one of the few studies in Tunisia to document HAI trends over a prolonged period spanning nearly a decade, using a standardized and reproducible methodology across five successive surveys—a design that confers particular reliability to the temporal trends reported. However, some limitations should be acknowledged. The PPS methodology, although well-suited for repeated trend monitoring, provides a cross-sectional snapshot that may not fully capture the true HAI incidence. Moreover, the single-center design limits generalizability.

Taken as a whole, these findings paint an alarming picture of a growing HAI burden calling for strengthened IPC programs targeting modifiable risk factors, sustained investment in microbiological surveillance, and a coordinated national response. Efforts to improve infection prevention in Tunisian hospitals are real and ongoing, as reflected in the successive implementation of IPC measures and the commitment to repeated prevalence monitoring over nearly a decade, achievements that deserve recognition. However, the sustained upward trend in HAI prevalence underscores that these efforts, although meaningful, remain insufficient and must be reinforced. In this perspective, building a national HAI surveillance network would be a pivotal milestone, providing the continuous, multi-site data needed to guide targeted interventions and antimicrobial stewardship.