Narrative Review of the Epidemiology of Hospital-Acquired Pneumonia and Ventilator-Associated Pneumonia in Gulf Cooperation Council Countries

Abstract

Hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP) are the most common healthcare-associated infections, with rates varying between countries. Antimicrobial resistance (AMR) among common HAP/VAP pathogens has been reported, and multidrug resistance (MDR) is of further concern across Middle Eastern countries. This narrative review summarizes the incidence and pathogens associated with HAP/VAP in hospitals across Gulf Cooperation Council (GCC) countries. A PubMed literature search was limited to available data on HAP or VAP in patients of any age published within the past 10 years. Reviews, non-English language articles, and studies not reporting HAP/VAP data specific to a GCC country were excluded. Overall, 41 articles, a majority of which focused on VAP, were selected for inclusion after full-text screening. Studies conducted over multiple years showed a general reduction in VAP rates over time, with Gram-negative bacteria the most commonly reported pathogens. Gram-negative isolates reported across GCC countries included Acinetobacter baumannii, Pseudomonas aeruginosa, and Klebsiella pneumoniae. Rates of AMR varied widely among studies, and MDR among A. baumannii, K. pneumoniae, Escherichia coli, P. aeruginosa, and Staphylococcus aureus isolates was commonly reported. In Saudi Arabia, between 2015 and 2019, rates of carbapenem resistance among Gram-negative bacteria were 19–25%; another study (2004–2009) reported antimicrobial resistance rates in Acinetobacter species (60–89%), P. aeruginosa (13–31%), and Klebsiella species (100% ampicillin, 0–13% other antimicrobials). Although limited genotype data were reported, OXA-48 was found in ≥ 68% of patients in Saudi Arabia with carbapenem-resistant Enterobacteriaceae infections. Ventilator utilization ratios varied across studies, with rates up to 0.9 reported in patients admitted to adult medical/surgical intensive care units in both Kuwait and Saudi Arabia. VAP remains a burden across GCC countries albeit with decreases in rates over time. Evaluation of prevention and treatment measures and implementation of a surveillance program could be useful for the management of HAP and VAP.

Key Summary Points

Hospital-acquired and ventilator-associated pneumonia (HAP/VAP) infections represent potentially costly comorbidities, with rates of acquisition that vary widely between nations.

Articles described VAP rates in Gulf Cooperation Council member states that varied widely by nation and time, with multi-year studies suggesting a trend of decreasing VAP rates over the last 10–15 years.

The most commonly reported organisms associated with HAP/VAP infections were multidrug-resistant species in the genera Acinetobacter, Pseudomonas, and Klebsiella, although methicillin (meticillin)-resistant Staphylococcus aureus and Streptococcus pneumonia were also commonly identified.

The high frequency of treatment-resistant bacterial pathogens represents a challenge to further improvement in managing these infections.

Introduction

Hospital-acquired infections (HAIs), also referred to by the broader term of healthcare-associated infections, are typically defined as infections that occur during care in a hospital facility that were not present or incubating at the time of admission [1]. HAIs usually manifest ≥ 48 h after hospital admission and include infections that present after hospital discharge. Hospital-acquired pneumonia (HAP; pneumonia occurring ≥ 48 h after hospital admission) and ventilator-associated pneumonia (VAP; pneumonia occurring ≥ 48 h after endotracheal intubation) are among the most frequent types of HAI both in high-income and low-/middle-income countries [1, 2]. Use of the term HAP varies across guidelines and studies [2]; in this review, HAP encompasses any pneumonia occurring ≥ 48 h after hospital admission and therefore includes VAP as a subset.

In general, there is wide variation in VAP rates reported across studies due to a combination of factors, including country, intensive care unit (ICU) type, patient case-mix, and differing definitions of VAP [3, 4]. Surveillance of device-associated HAIs during 2012–2017 in ICUs across 45 countries of the International Nosocomial Infection Control Consortium [INICC; predominantly countries with developing economies, including Bahrain, Kuwait, Saudi Arabia, and the United Arab Emirates (UAE)] determined pooled mean VAP rates for 11 ICU types [5]. VAP rates ranged from 7.4 (surgical cardiothoracic ICUs) up to 17.7 per 1000 ventilator-days (medical cardiac ICUs), which were considerably higher than those reported in the USA.

The pathogens most frequently associated with VAP include Pseudomonas aeruginosa, Acinetobacter baumannii, Gram-negative bacilli of the Enterobacterales family (particularly Escherichia coli and Klebsiella pneumoniae), and Staphylococcus aureus [2, 3]. Antimicrobial resistance rates, including multidrug resistance among common HAP/VAP pathogens, are substantial. International surveillance data report methicillin (meticillin) resistance in approximately one half of S. aureus isolates, high rates of cefepime and piperacillin-tazobactam resistance in P. aeruginosa (up to 30%), and carbapenem resistance in > 50% of A. baumannii isolates [2]. Of further concern, multidrug resistant (MDR) pathogens pose a serious problem in Middle Eastern countries due to multiple predisposing factors, including antibiotic misuse and overuse, continuous population movement, and inadequate infection prevention and control strategies [6]. Other inherently MDR organisms as a result of extensive carbapenem use (namely Stenotrophomonas maltophilia) are also encountered in critical care units. Furthermore, apart from bacteria, certain fungi (in particular Candida auris) can be challenging for hospitals and clinicians in view of MDR and their propensity to spread, persist in the hospital environment, and cause outbreaks in ICUs that are difficult to treat [7, 8].

Infection control and surveillance in Gulf Cooperation Council (GCC) countries is coordinated by the GCC Centre for Infection Control (GCC-CIC) [9]. In addition to providing evidence-based guidelines for infection control practices, the GCC-CIC conducts educational and training activities for personnel working in infection prevention and control [9]. Current standards for respiratory therapy procedures are contained in the GCC Infection Prevention and Control Manual, including detailed protocols for the setup and maintenance of mechanical ventilation circuits and artificial airways, specimen collection, personal protective equipment, and hand hygiene [10]. However, further initiatives and more extensive participation in training activities for infection surveillance and data management are needed, and there is not yet a comprehensive and integrated system in place for collecting and compiling data from hospitals throughout the region [9].

Although all GCC countries are classified as high-income, their economies are still developing, and their healthcare systems face many of the challenges common to other developing nations [9, 11]. Subsequently, this narrative review was undertaken to summarize available data related to the incidence and associated organisms of HAP and VAP in hospitals throughout GCC countries, with a view to identifying unmet medical needs and future challenges for the management of HAP/VAP in the region.

Methods

A PubMed database search was carried out on 13 November 2021 using the search terms: “pneumonia, ventilator associated”[MeSH Terms] OR (pneumonia AND “ventilator associated”) OR “ventilator-associated pneumonia” OR (ventilat* AND associated AND pneumonia) OR ((device OR catheter) AND associated AND pneumonia) OR “healthcare-associated pneumonia”[MeSH Terms] OR (“healthcare associated” AND pneumonia) OR “healthcare-associated pneumonia” OR (healthcare AND associated AND pneumonia) OR “hospital-acquired pneumonia” OR (“hospital acquired” AND pneumonia) OR (nosocom* AND pneumonia) AND (Gulf OR Bahrain OR Kuwait OR Oman OR Qatar OR “Saudi Arabia” OR “United Arab Emirates” OR UAE) NOT Review[publication type]. The search was additionally limited to English-language articles published in the preceding 10 years.

Search results were reviewed by the authors to identify articles reporting original data related to HAP or VAP in patients of any age; articles that did not report HAP or VAP data specific to a GCC country were excluded. Search results were initially screened for relevance on the basis of the title and abstract, then full copies of each identified article were further reviewed for relevance.

This article is based on previously conducted studies and does not contain any new studies with human participants performed by any of the authors.

Results

Based on title and abstract, 57 articles were identified for full-text review. Among the full-text articles assessed, 41 were ultimately selected for potential inclusion in the review (Fig. 1), including studies conducted in Saudi Arabia (n = 32), Kuwait (n = 4), Qatar (n = 2), Oman (n = 1), UAE (n = 1), and Bahrain, Oman, and Saudi Arabia (n = 1). The majority of identified studies focused on VAP, either exclusively or in the context of HAIs, with only 2 publications [12, 13] focusing on HAP; thus, the majority of the discussion in the current review focuses on VAP.

Fig. 1.

Selection process. GCC Gulf Cooperation Council, HAP hospital-associated pneumonia, VAP ventilator-associated pneumonia

VAP Rates

Annual VAP rates varied widely between studies and over time (Fig. 2). In the studies that reported data for multiple years, there was a general trend for reduced VAP rates over time. In support of this observation, a longitudinal study (2008–2013) conducted in Bahrain, Oman, and Saudi Arabia showed a general trend for reduced ICU-associated VAP rates over time (Fig. 3A) [9]. A number of time-series analyses reported reduced VAP rates with VAP prevention bundles [9, 14–19]. Studies identified a need for further study of prevention and intervention programs, evidence-based treatment guidelines, and increased VAP-related education activities for staff to ensure that protocols are properly implemented [20–22].

Fig. 2.

VAP rates in adult ICUs across GCC countries as reported by various studies included in this review [9, 14, 17–19, 23, 29, 32, 34]. The inset shows ICU type for data from work by Ali et al. (2016), which was the only included study describing rates in a single GCC nation, Qatar [23]. GCC Gulf Cooperation Council, ICU intensive care unit, VAP ventilator-associated pneumonia

Fig. 3.

VAP rates across ICUs in Bahrain, Oman, and Saudi Arabia, 2008–2013, as reported by El-Saed 2016 [9] A by year and B by ICU type. Note that, for panel B, the VAP rate for trauma ICUs is reported for the time period 2010–2013 (not 2008–2013). ICU intensive care unit, VAP ventilator-associated pneumonia

One study from Qatar from 2010 to 2012 that analyzed data by ICU type demonstrated higher VAP rates within trauma ICUs than within medical or surgical ICUs (Fig. 2, inset) [23]. In contrast, a study from Bahrain, Oman, and Saudi Arabia from 2008 to 2013 reported higher VAP rates within medical/surgical ICUs than trauma ICUs, although data from trauma ICUs were collected only from 2010 to 2013 (Fig. 3B) [9].

Infective Agents and Antimicrobial Resistance

Table 1 summarizes the reported associated organisms of HAP/VAP and antimicrobial resistance data. It is noteworthy that isolation of organisms from respiratory samples is not evidence that they are causative of the HAP/VAP unless they are isolated from both the blood culture and respiratory samples at the same time, which is uncommon. Thus, for the purposes of this review, we refer to organisms as being associated with VAP.

Table 1.

HAP-/VAP-associated organisms and antimicrobial resistance

Article	Study period	Study population and setting	ICU type	Isolated organisms	Antimicrobial resistance
Saudi Arabia
Almangour et al., 2021 [45]	May 2015–May 2019	Adults ≥ 18 years treated at two hospitals in Riyadh who developed VAP due to MDR Gram-negative pathogens and who were treated with aerosolized plus IV colistin (n = 65) or IV colistin alone (n = 70)	NOS	MDR Gram-negative  MDR P. aeruginosa (51%)  MDR A. baumannii (35%)  MDR K. pneumoniae (12%)	No colistin-resistant isolates were identified
Mahmoud et al., 2021 [25]	Jul 2016–Sep 2019	Adults > 18 years classified as “comfort care” or “support care” patients and required an infection-related admission to KAMC hospital (Riyadh) during their first year	NA	Gram-negative  Pseudomonas spp., n = 58 (17%)  Acinetobacter spp., n = 28 (8%)	MDR, 92 of 342 (27%); Carbapenem resistance, 65 of 342 (19%)
				Gram-positive  S. aureus, n = 45 (13%)
Osman et al., 2020 [31]	Jan 2015–Mar 2018 (pre-bundle, Jan 2015–Feb 2017; post-bundle, Mar 2017–Mar 2018)	Patients aged 1 month to 14 years admitted to pediatric ICU of KAMC hospital, Jeddah, and requiring MV for > 48 h (n = 131)	Pediatric	Gram-negative (pre-bundle; post-bundle)  P. aeruginosa (36%; 27%)  K. pneumoniae (12%; 36%)  S. maltophilia (15%; 9%)  A. baumannii (12%; 9%)  E. cloacae (9%; 9%)  E. coli (6%; 9%)  H. influenzae (6%; 0%)  Moraxella catarrhalis (3%; 0%)  Serratia marcescens (3%; 0%) Gram-positive  S. aureus (9%; 0%)  S. pneumoniae (6%; 0%)	Carbapenem-resistant Gram-negative, 11 of 44 (25%)  Pre-bundle group (26%)  Post-bundle group (18%)
Alsulami et al., 2020 [46]	Nov 2014–Jun 2019	Patients aged 14–84 years who underwent open heart surgery with cardiopulmonary bypass in the surgery department of a single hospital, Jeddah (n = 200)	Surgical	Gram-negative  H. influenzae, n = 3	NR
Alraddadi et al., 2019 [13]	Jan–Nov 2017; Dec 2017–Aug 2018 (date ranges were compared since CAZ-AVI was not available until Dec 2017)	Patients > 18 years at a single hospital in Jeddah who had CRE infections:  Treated with CAZ-AVI for ≥ 24 h (total n = 10; HAP n = 5)  Treated with other agents (total n = 28; HAP n = 14)	NA	Gram-negative, CAZ-AVI; comparator  K. pneumoniae, 7 of 10 (70%); 23 of 28 (82%)  E. coli, 3 of 10 (30%); 5 of 28 (18%)	Carbapenem resistance (100%)
Alshamrani et al., 2019 [47]	May 2017	All inpatients of all ages at six hospitals in five cities	NA	Gram-negative  Pseudomonas spp. (19%)  Klebsiella spp. (19%)  E. coli (13%)  Acinetobacter spp. (7%)	NR
				Gram-positive  S. aureus (7%)
Alyami et al., 2020 [48]	Jun 2013–May 2019	Patients of all ages infected or colonized by Chryseobacterium/Elizabethkingia spp. at Prince Sultan Military Medical City, Riyadh	NOS	Gram-negative  E. meningoseptica, 22 of 27 (81%)  C. indologenes, 4 of 27 (15%)  Other Chryseobacterium/Elizabethkingia spp., 1 of 27 (4%)	NR
Balkhy et al., 2020 [26]	2008–2016	Patients of all ages diagnosed with DA-HAIs in ICUs of Ministry of National Guard Health Affairs (MNGHA) hospitals in Riyadh, Jeddah, Alhassa, and Dammam	NOS	Gram-negative  Pseudomonas spp., 20 of 76 (26%)  Acinetobacter spp., 18 of 76 (24%)  Klebsiella spp., 10 of 76 (13%)  E. coli, 6 of 76 (8%)  S. maltophilia, 4 of 76 (5%)  Enterobacter spp., 3 of 76 (4%)  Serratia spp., 2 of 76 (3%)  Other, 6 of 76 (8%)	MRSA (50%) MDR  Acinetobacter spp. (89%)  Pseudomonas spp. (25%)  Klebsiella spp. (11%)  Enterobacter spp. (33%)  E. coli (17%)  Serratia spp. (0%)  Stenotrophomonas spp. (0%)
				Gram-positive   S. aureus, 2 of 76 (3%)
				Fungi   Candida spp., 4 of 76 (5%)  Non-candidal yeast, 1 of 76 (1%)
Bosaeed et al., 2020 [28]	Jan 2017–Dec 2018	Patients ≥ 18 years diagnosed with MDR P. aeruginosa infection and treated with ceftolozane-tazobactam for ≥ 72 h at the KAMC, Riyadh	NOS	MDR Gram-negative  P. aeruginosa, n = 19	MDR P. aeruginosa (HAP/VAP cases):  Ceftazidime and ciprofloxacin resistant, 1 of 3 (33%)  Amikacin resistant, 1 of 3 (33%)
Hala et al., 2019 [49]	Nov 2017	Adult male patient (mid-60s) admitted with trauma-associated airway obstruction who underwent a percutaneous tracheostomy and developed infection 1 week after surgery	NA	Gram-negative  Klebsiella quasipneumoniae subspecies similipneumoniae	CRE
Abdallah et al., 2018 [12]	Feb–Mar 2017	Male aged 31 years admitted to ICU post-exploratory laparotomy/thoracotomy with associated chest and laparotomy infections (and further complications of septic shock, acute kidney injury, and internal fixation of broken elbow)	NA	Gram-negative  P. stuartii  K. pneumoniae  A. baumannii	CRE  P. stuartii  K. pneumoniae MDR  A. baumannii
Al-Abdely et al., 2018 [15]	Sep 2013–Feb 2017	37 adult ICUs in 22 hospitals across 14 cities in Saudi Arabia	NOS	Gram-negative, baseline; intervention  A. baumannii, 46 of 110 (42%); 169 of 405 (42%)  K. pneumoniae, 14 of 110 (13%); 51 of 405 (13%)  P. aeruginosa, 13 of 110 (12%); 61 of 405 (15%)  Acinetobacter spp., 8 of 110 (7%); 21 of 405 (5%)  E. coli, 3 of 110 (3%); 14 of 405 (3%)  S. maltophilia, 0 of 110 (0%); 9 of 405 (2%)	NR
				Gram-positive, baseline; intervention  S. aureus, 11 of 110 (10%); 28 of 405 (7%)
				Fungi, baseline; intervention  C. albicans, 6 of 110 (5%); 4 of 405 (1%)  Candida spp., 1 of 110 (1%); 7 of 405 (2%)
Alessa et al., 2018 [50]	NR	Male African-American nursing home resident aged 79 years (with multiple comorbidities, including end-stage renal disease, receiving intermittent hemodialysis and with history of P. aeruginosa airway colonization) who presented to hospital with shortness of breath and potential nasogastric tube misplacement/suspected aspiration pneumonia	NA	Gram-negative  P. aeruginosa	MDR  P. aeruginosa
Khan et al., 2016 [51]	Aug 2003–Dec 2010	All patients on MV for > 48 h in the KAMC tertiary care hospital adult ICU, Riyadh (patients with burns, brain death, do-not-resuscitate orders, or transferred from other hospitals were excluded)	Adult	Gram-negative (EO-VAP; LO-VAP)  H. influenzae (25%; 2%)  P. aeruginosa (5%; 16%)  A. baumannii (13%; 18%)  K. pneumoniae (10%; 5%)  Enterobacter spp. (6%; 6%)	NR
				Gram-positive (EO-VAP; LO-VAP)  MSSA (11%; 7%)  MRSA (5%; 4%)  S. pneumoniae (8%; 2%)
Al-Obeid et al., 2015 [33]	Jan–Dec 2012	A. baumannii isolates collected from patients at the Security Forces Hospital During 2012, 12 XDR- A. baumannii strains were isolated from tracheal samples of ICU patients with serious VAP (mean age 59.2 years)	NOS	Gram-negative  A. baumannii (100%)	XDR- A. baumannii (> 90%)
Shaath et al., 2014 [52]	Mar–Sep 2010	Postoperative cardiac patients aged < 14 years in the pediatric cardiac ICU of KAMC hospital, Riyadh	Pediatric	Gram-negative (33%)	NR
Balkhy et al., 2014 [27]	Oct 2004–Jun 2009	Adult ICU of KAMC tertiary care hospital, Riyadh	Adult	Gram-negative  Acinetobacter (35%)  P. aeruginosa (25%)  Klebsiella spp. (6%)  Enterobacter spp. (4%)  Haemophilus spp. (4%)  S. maltophilia (3%)  E. coli (2%)	MDR  Acinetobacter spp. (60–89%)   3-class (86%)   4-class (69%) P. aeruginosa (13–31%)   3-class (13%)   4-class (10%) S. aureus (oxacillin 42%) Coagulase-neg staphylococci (VRE 17%; oxacillin 100%)
				Gram-positive  S. aureus (including MRSA; 17%)  Coagulase-neg staphylococci (3%)
El-Saed et al., 2013 [53]	Aug 2003–Jun 2009	Adult general ICU of KAMC tertiary care hospital, Riyadh	Adult/general	Gram-negative  Acinetobacter spp. (27%)  P. aeruginosa (22%)  Klebsiella spp. (7%)  Haemophilus spp. (6%)  Enterobacter spp. (5%)	NR
				Gram-positive  S. aureus (15%; 37 of 70 were MRSA)
Bukhari et al., 2012 [19]	Jan–Dec 2010	Patients admitted to the adult medical/surgical ICU at Hera General Hospital, Makkah	Adult	Gram-negative  P. aeruginosa (31%)  A. baumannii (28%)  K. pneumoniae (14%)  E. coli (8%)	MDR  A. baumannii (100%)  P. aeruginosa (93%)  K. pneumoniae (83%)  MRSA (15%)
				Gram-positive  MRSA (15%)  S. aureus (2%)
Al-Dorzi et al., 2012 [14]	Aug 2003–Jun 2009	Patients admitted to the adult ICU of KAMC hospital in Riyadh who required MV (burn patients and brain death patients were excluded)	General	Gram-negative, 251 of 327 (77%)	NR
				Gram-positive, 60 of 327 (20%)
Kuwait
Alfouzan et al., 2021 [29]	2018, 2019	Patients ≥ 18 years admitted to a surgical/medical ICU at Farwania Hospital, a government secondary-care hospital in Kuwait	General	Gram-negative  A. baumannii, 9 of 30 (30%)  K. pneumoniae, 7 of 30 (23%)  P. aeruginosa, 3 of 30 (10%)  E. coli, 2 of 30 (7%)  Enterobacter cloacae, 2 of 30 (7%)  Morganella morganii, 1 of 30 (3%)  Citrobacter koseri, 1 of 30 (3%)	NR (not specified in patients with HAP/VAP)
				Gram-positive  Methicillin-susceptible S. aureus, 1 of 30 (3%)
Abulhasan et al., 2020 [32]	Jan 2015–Dec 2017	Patients admitted to the Neurocritical Care Unit of Ibn Sina Hospital, a tertiary care teaching hospital in Kuwait (97% of patients included in the study were aged ≥ 18 years)	Neurocritical	Gram-negative  Klebsiella spp., 11 of 33 (33%)  E. coli, 4 of 33 (12%)  P. aeruginosa, 4 of 33 (12%)  Enterobacter, 5 of 33 (15%)  A. baumannii, 1 of 33 (3%)  S. maltophilia, 1 of 33 (3%)  Other Enterobacteriaceae, 1 of 33 (3%)	NR (not specified for HAP patients)
				Gram-positive  S. aureus, 5 of 33 (15%)  MRSA, 1 of 33 (3%)
Al-Mousa et al., 2018 [16]	Jan 2014–Mar 2015 (baseline period, Jan–Mar 2014; intervention period, Apr 2014–Mar 2015)	Patients admitted to three adult ICUs in two hospitals in Kuwait City	General	Gram-negative (baseline; intervention)  A. baumannii (58%; 59%)  P. aeruginosa (8%; 13%)  Stenotrophomonas spp. (8%; 6%)  K. pneumoniae (8%; 0%)  Serratia marcescens (8%; 0%)  Enterobacter spp. (0%; 6%)  E. coli (0%; 3%)  Morganella morganii (0%; 3%)	NR
				Gram-positive  S. aureus (8%; 0%)
				Fungi  Candida spp. (0%, 9%)
Oman, Qatar, or UAE
Sannathimmappa et al., 2021 [30]	Jan 2017–Aug 2019	All bacterial isolates from ET aspirates of ventilated patients (> 95% of sampled patients were aged ≥ 21 years) at a tertiary care ministry hospital in the North-Batinah region, Oman	NOS	Gram-negative  A. baumannii (31%)  K. pneumoniae (24%)  P. aeruginosa (23%)  E. coli (3%) Other Enterobacteriaceae (5%)  S. maltophilia (2%)	MDR  A. baumannii (86%)  K. pneumoniae (73%)  E. coli (67%)  P. aeruginosa (24%)  S. aureus (53%)  Enterococcus spp. (50%) ESBL  E. coli, 3 of 4 (75%)  K. pneumoniae, 6 of 35 (17%)  Carbapenem resistance  K. pneumoniae, 23 of 35 (66%)
				Gram-positive  S. aureus (4%)  MRSA (5%)  Streptococcus spp. (2%)  Enterococcus spp. (1%)
Ali et al., 2016 [23]	Jan 2010–Dec 2012	All patients ≥ 15 years clinically diagnosed with VAP in three adult ICUs (medical, surgical, and trauma) of Hamad General Hospital, Qatar	General	Gram-negative  Pseudomonas spp., 39 of 106 (37%)  Klebsiella spp., 25 of 106 (24%)  Enterobacter spp., 24 of 106 (23%)  Acinetobacter spp., 23 of 106 (22%)  Haemophilus spp., 16 of 106 (15%)  E. coli, 5 of 106 (5%)  Stenotrophomonas spp., 4 of 106 (4%)	MDR, 43 of 106 (41%)
				Gram-positive  Staphylococcus spp., 21 of 106 (20%)  Streptococcus spp., 5 of 106 (5%)
Arumugam et al., 2018 [54]	Jan 2010–Jan 2013	Adult trauma patients admitted to the Trauma Centre of Hamad General Hospital who required intubation (either before hospitalization or in the trauma room) and ventilation (patients with burns, drowning, death, or discharged or those who were transferred to other facilities < 48 h after admission were excluded)	Trauma	Gram-negative (All VAP cases; VAP in PHI; VAP in TRI)  K. pneumoniae (36%; 22%; 38%)  H. influenzae (30%; 30%; 29%)  E. cloacae (12%; 9%; 15%)  P. aeruginosa (12%; 9%; 15%)  A. baumanii (7%; 4%; 9%)  Moraxella catarrhalis (5%; 4%; 6%)  K. oxytoca (4%; 4%; 3%)  K. ozonae (5%; 13%; 0%)	MDR  A. baumannii, 1 of 23 (4%) patients intubated before hospitalization  A. baumannii and S. aureus, 2 of 34 (6%) patients intubated on arrival in the trauma room
				Gram-positive  S. aureus (28%; 35%; 24%)  S. pneumoniae (16%; 13%; 18%)
Sonnevend et al., 2013 [55]	2009–2011	Clinically relevant NDM-producing carbapenem-resistant Enterobacteriaceae isolated from major hospitals in the UAE	NA	Gram-negative  K. pneumoniae, 3 of 5 (60%)  E. coli, 2 of 5 (40%)	Carbapenem resistance  K. pneumoniae (100%)  E. coli (100%)

CAZ-AVI ceftazidime-avibactam, CRE carbapenem-resistant Enterobacteriaceae, DA device-associated, ESBL extended-spectrum β-lactamase, EO early onset, ET endotracheal, HAI hospital-acquired infection, HAP hospital-acquired pneumonia, ICU intensive care unit, IV intravenous, KAMC King Abdulaziz Medical City, LO late onset, MDR multidrug-resistant, MIC minimum inhibitory concentration, MRSA methicillin-resistant S. aureus, MSSA methicillin-sensitive S. aureus, MV mechanical ventilation, NA not applicable, NDM New Delhi metallo-beta-lactamase, NOS not otherwise specified, NR not reported, PHI prehospital intubation, TRI trauma room intubation, UAE United Arab Emirates, VAP ventilator-associated pneumonia, XDR extensively drug-resistant

Gram-negative bacteria were the most commonly reported organisms overall, including Acinetobacter species (particularly A. baumannii), Pseudomonas species (particularly P. aeruginosa), and Klebsiella species (particularly K. pneumoniae). Gram-positive bacteria included methicillin-resistant S. aureus and Streptococcus pneumoniae. Other pathogens of note were S. maltophilia, which is an important nosocomial pathogen in children in Saudi Arabia [24], and carbapenem-resistant Providencia stuartii [12] (Table 1).

Among studies that reported antimicrobial resistance, rates of resistance varied widely. Multidrug resistance (i.e., resistance to ≥ 1 agent in ≥ 3 antimicrobial classes) was frequently reported among A. baumannii, K. pneumoniae, E. coli, P. aeruginosa, and S. aureus isolates [23, 25–30]; vancomycin-resistant Enterococcus [26] and MDR Enterococcus species [26, 30] were rarely reported. A study in Saudi Arabia conducted from 2015 to 2018 reported carbapenem-resistant Gram-negative organisms in 25% of all VAP cases [31], while another study from Saudi Arabia from 2016 to 2019 reported 19% [25]. Two studies in Kuwait, one in Qatar, and one in Oman reported extended-spectrum beta-lactamase K. pneumoniae (17–32%), E. coli (75%), or Enterobacterales isolates (24%) [23, 29, 30, 32]. From 2004 to 2009, a tertiary care hospital in Riyadh reported 60–89% resistance to all tested antimicrobials for Acinetobacter species and 13–31% for P. aeruginosa; Klebsiella species were fully resistant to ampicillin and 0–13% were resistant to other tested antimicrobials [27].

Genotype data were sparse in the reviewed studies. One Saudi Arabian study that analyzed 12 extensively drug-resistant (XDR; defined as resistant to at least four classes of antimicrobials, including carbapenems) A. baumannii isolates from patients with VAP reported that none of the 12 isolates harbored a class B carbapenemase gene; class A carbapenemase bla_GES was detected in 2 out of 12 isolates. Insertion sequence ISAba1 was detected in 5 out of 12 isolates, all isolates harbored Acinetobacter-derived cephalosporinases, and all isolates were carbapenem resistance–associated OM protein negative (i.e., had no evidence of efflux-mediated carbapenemase resistance) [33]. In another study in Saudi Arabia in patients with carbapenem-resistant infections, OXA-48 was the predominant carbapenemase in ≥ 68% of patients [13].

Ventilator Utilization Ratios

Table 2 shows reported VAP rates and ventilator utilization ratios, which varied widely over time between ICU type and between studies. A study in Kuwaiti patients from 2013 to 2015 reported a utilization rate of 0.9 for adult medical/surgical ICUs and 0.1 for adult coronary ICUs [34]. Another study in Saudi Arabia reported utilization rates of 0.7, 0.6, and 0.2 for adult medical, surgical, and cardiac ICUs, respectively [17], while a study from 2008 to 2013 conducted in Bahrain, Oman, and Saudi Arabia reported rates of 0.5–0.6 across adult ICUs [9]. There was no discernable trend over time in that study, with reported utilization ratios of 0.5 in 2008, 0.6 in 2012, and 0.5 in 2013 [9]. Another study in Saudi Arabia reported a decrease in utilization rates from 0.7 in 2003 to 0.5 in 2009 [14], while a study in Kuwait reported a rate increase from 0.7 to 0.8 in 2014 to 2015 [16].

Table 2.

Reported VAP rates and ventilator utilization ratios—all ICU types

Source	City/country	ICU type	Year(s)	VAP definition used	VAP ratea	Ventilator utilization ratiob
Al-Dorzi et al., 2012 [14]	Riyadh, Saudi Arabia	Adult medical, surgical, trauma	2003	US CDC criteria	19.1	0.7
			2009		6.3	0.5
			2003–2009		15.9	0.6
Bukhari et al., 2012 [19]	Makkah, Saudi Arabia	Adult medical, surgical	2009	Pneumonia in a patient intubated and ventilated at the time of, or ≤ 48 h before the onset of the event	3.4	NR
			2010		2.0	NR
Al-Tawfiq et al., 2013 [17]	Dhahran, Saudi Arabia	Adult medical, surgical, cardiac	2004	MV with chest X-ray showing new or progressive infiltrate, consolidation, cavitation, or pleural effusion	9.8	NR
			2011		1.3	NR
			2004–2011		4.5	NR
		Adult medical	2004–2011		5.0 (0.4–9.4)	0.7 (0.6–0.7)
		Adult surgical	2004–2011		4.9 (2.1–8.4)	0.6 (0.5–0.7)
		Adult cardiac	2004–2011		1.9 (0–3.7)	0.2 (0.2–0.3)
Al-Thaqafy et al., 2014 [18]	Riyadh, Saudi Arabia	Adult medical, surgical, trauma	2010	US CDC/NHSN criteria	3.6	0.7
			2013		1.0	0.6
Shaath et al., 2014 [52]	Riyadh, Saudi Arabia	Pediatric cardiac (postcardiac surgery patients)	2010	US CDC criteria	29.0	0.3
Ali et al., 2016 [23]	Doha, Qatar	Adult medical, surgical, trauma	2010	US CDC criteria	5.4	NR
			2011		5.9	NR
			2012		3.9	NR
		Adult medical	2010		3.9	NR
			2011		4.0	NR
			2012		2.3	NR
		Adult surgical	2010		3.1	NR
			2011		5.7	NR
			2012		5.1	NR
		Adult trauma	2010		10.1	NR
			2011		8.7	NR
			2012		5.8	NR
Al-Mousa et al., 2016 [34]	Kuwait City, Kuwait	Adult medical, surgical	Nov 2013–Mar 2015	US CDC/NHSN criteria	4.0 (2.9–5.3)	0.9 (0.9–0.9)
		Adult coronary	Nov 2013–Mar 2015		12.9	0.1 (0.1–0.1)
		Pediatric	Nov 2013–Mar 2015		0.3 (0.2–4.6)	0.7 (0.7–0.7)
		Neonatal	Nov 2013–Mar 2015		1.0	0.5 (0.4–0.5)
El-Saed et al., 2016 [9]	Bahrain, Oman, and Saudi Arabia	Medical/surgical, neurosurgical, surgical, trauma, pediatric, pediatric cardiothoracic, and neonatal	2008–2013	NHSN criteria	4.8 (4.3–5.3)	0.6 (0.6–0.6)
			2008		7.4 (5.5–9.2)	0.5 (0.5–0.5)
			2009		6.6 (5.1–8.1)	0.6 (0.6–0.6)
			2010		4.7 (3.7–5.7)	0.6 (0.6–0.6)
			2011		5.4 (4.3–6.5)	0.6 (0.6–0.6)
			2012		3.0 (2.1–3.8)	0.6 (0.6–0.6)
			2013		2.0 (0.9–3.0)	0.5 (0.5–0.5)
		Medical/surgical	2008–2013		5.4 (4.8–6.0)	0.6 (0.6–0.6)
		Pediatric	2008–2013		2.4 (0.8–3.9)	0.5 (0.5–0.5)
		Neonatal	2008–2013		1.8 (0.9–2.8)	0.3 (0.3–0.3)
		Trauma	2010–2013		1.3 (0.0–3.9)	0.6 (0.6–0.6)
		Neurosurgical	2010		4.3 (0.1–8.5)	0.6 (0.6–0.6)
		Pediatric cardiothoracic	2010		3.5 (0.1–7.0)	0.6 (0.6–0.6)
		Surgical	2013		1.6 (0.0–4.8)	0.5 (0.4–0.5)
Banjar et al., 2017 [20]	Makkah region, Saudi Arabia	Adult, surgical, coronary care, pediatric, and neonatal	2013	NR	6.9c (IQR, 3.6–10.3)	NR
	Makkah, Makkah, Saudi Arabia	Not specified	2013		6.4c (IQR, 3.2–8.2)	NR
	Jeddah, Makkah, Saudi Arabia	Not specified	2013		8.8c (IQR, 4.6–20.2)	NR
	Taif, Makkah, Saudi Arabia	Not specified	2013		6.1c (IQR, 3.6–11.6)	NR
	Qunfudah, Makkah, Saudi Arabia	Not specified	2013		4.4c (IQR, 0.6–6.3)	NR
Al-Mousa et al., 2018 [16]	Kuwait City, Kuwait	Adult medical/surgical, coronary	Jan–Mar 2014	US CDC/NHSN criteria	7.0 (3.8–11.8)	0.7
			Apr 2014–Mar 2015		3.6 (2.5–5.0)	0.8
Gaid et al., 2018 [56]	Provinces of Asser, Jeddah, Riyadh, and Qassim, Saudi Arabia	Adult medical/surgical	May 2015–Feb 2016	NR	18.1–26.6	0.5 (0.5–0.6)
	Provinces of Asser and Taif, Saudi Arabia	Adult medical/surgical	Sep 2013–Mar 2015		9.3–20.7	0.3 (0.3–0.3)
	Provinces of Najran and Tabuk, Saudi Arabia	Adult medical/surgical	Sep 2013–Feb 2016		0.9–16.4	0.7 (0.6–0.7)
	Provinces of Taif, Hail, and Madina, Saudi Arabia	Adult medical/surgical	Sep 2015–Mar 2016		10.1–51.6	0.9 (0.9–1.0)
	Province of Riyadh, Saudi Arabia	Adult medical/surgical	Jan 2015–Feb 2016		186.5	0.1 (0.04–0.1)
Abulhasan et al., 2020 [32]	Kuwait	Neurocritical care (97% adult patients)	2015	US CDC/NHSN criteria	11.1	0.4
			2016		5.7	0.2
			2017		6.1	0.4
Ahmed et al., 2021 [57]	Riyadh, Saudi Arabia	Not specified (60-bed hospital)	2019	US CDC criteria	2.1	NR
Alfouzan et al., 2021 [29]	Kuwait	Adult medical/surgical	2018, 2019	Pneumonia in a patient on MV for > 2 days on the date of event	4.2	0.7

CI confidence interval, ICU intensive care unit, IQR interquartile range, MV mechanical ventilation, NHSN National Healthcare Safety Network, NR not reported, US CDC US Centers of Disease Control and Prevention, VAP ventilator-associated pneumonia

^aPer 1000 ventilator-days (where provided, error is 95% CI, unless otherwise indicated)

^bVentilator utilization ratio is the number of ventilator days/number of patient days (where provided, error is 95% CI)

^cMedian value

Discussion

This narrative review was undertaken to summarize available data related to the incidence and causative organisms of HAP and VAP in hospitals throughout GCC countries. Data from multi-year studies suggest that VAP rates have generally decreased over time in GCC countries [16, 18, 19, 23, 32]. Although time-series analyses suggest that VAP prevention bundles reduce VAP rates [9, 14–19], the contribution of individual components is somewhat unclear. A recent review found no high-level evidence to support an association between ventilator bundle implementation and reduced ventilator-associated event (VAE) risk [3]. Indeed, findings suggest that chlorhexidine and stress ulcer prophylaxis may actually increase VAE risk, while reduced mechanical ventilation duration will likely reduce VAE risk [3]. In controlled studies, a significant impact of probiotics on VAP rate has not been observed. These studies include a randomized controlled study carried out in ICUs across North America and Saudi Arabia to assess the effect of the probiotic Lactobacillus rhamnosus GG on VAP development in critically ill patients. Results of the study showed no significant difference in VAP incidence between the probiotic and placebo groups, or for any of the other prespecified outcomes [35].

Additionally, our literature review suggests that the most common organisms associated with VAP in GCC countries were Gram-negative bacteria, particularly Acinetobacter species, Pseudomonas species, and Klebsiella species, as well as Enterococcus species, with MDR reported in many studies. The types of organisms associated with HAP/VAP and the proportions of positive isolates were similar between the countries studied (Saudi Arabia, Kuwait, and Oman; Qatar; or UAE). E. coli and P. aeruginosa were present in up to approximately 40% of isolates, and A. baumannii and K. pneumoniae were isolated from > 50% of isolates in at least one of the included studies.

Our analysis also found a wide range of ventilator utilization ratios, with higher ratios increasing the likelihood of developing VAP. Thus, ventilator utilization ratio may be a useful outcome measure in VAP prevention studies [18]. However, ventilator utilization ratios are necessarily higher in certain patient populations (e.g., very-low-birth-weight infants, patients with chronic obstructive pulmonary disease); therefore, it may be misleading to compare ratios across different hospitals/facilities due to differences in patient populations [36].

Infection prevention programs commonly incorporate multimodal horizontal strategies that aim to reduce risk from all nosocomial pathogens, such as hand hygiene protocols, environmental disinfection, and prevention bundles [37, 38]. Internationally, many organizations have developed VAP prevention guidelines and tools, and VAP bundles have been widely implemented in hospitals [39–41]. While infection control and surveillance in GCC countries is coordinated by the GCC-CIC and evidence-based guidelines for infection control practices have been provided by the organization [9], local differences in antimicrobial policies and antimicrobial stewardship programs may exist [42]. Substantial variability has been noted among the VAP bundles adopted by different hospitals in other countries, although utilized components frequently include head-of-bed elevation by 30–45°, daily interruption of sedation and assessment of readiness to extubate, use of endotracheal tubes with subglottic secretion drainage ports, and avoidance of ventilator circuit changes unless visible soiling is present [40, 41]. The INICC Multidimensional Approach for VAP rate reduction incorporates multiple strategies: a VAP prevention bundle; education; outcome and process surveillance; and provision of feedback related to VAP rates, VAP consequences, and performance [34]. VAP data also constitute part of the performance dashboard of hospital infection prevention and control committees.

A limitation of this analysis is that the data were compiled from different and diverse healthcare organizations and laboratories, which have variations in definitions and diagnostic limitations, as well as differing methodologies and technologies used in the detection, speciation, and susceptibility testing of isolated organisms from patients with VAP [3]. An illustration of this variation is the implementation of the VAE model by the US National Healthcare Safety Network (NHSN) in 2013 [43], which resulted in a large difference in reported VAP incidences between the USA and Europe [3]. Further evidence of these variations and the limitations they place on these data analyses was found in the surveillance of device-associated HAIs in ICUs across the INICC, which showed that pooled mean VAP rates ranged from 7.4 to 17.7 per 1000 ventilator-days depending on ICU type. These were considerably higher than 2012/2013 US NHSN pooled means (0.7–3.6 per 1000 ventilator-days) [5]. Accordingly, comparison of national VAP rates with the USA has been cautioned against due to varying classifications. Also, studies evaluating the effect of prevention bundles on VAP rates may be difficult to generalize to the overall population. This paper is a review of different retrospective studies, and thus, generalization of the impact of prevention bundles is difficult due to missing data since, generally, most papers were not studying prevention bundles and a few studies correlated bundle implementation with VAP rates. Another limitation of the current review is that all of the included studies were conducted before the onset of the coronavirus disease 2019 (COVID-19) pandemic and, therefore, do not take into consideration the likely increases in VAP rates [44].

Conclusions

Although VAP rates have generally decreased over time in GCC countries, VAP and associated MDR organisms remain a burden. Overall, there is a need for further research, including prospective, randomized, multicenter studies carried out in the Gulf Region, to assess the diagnosis, prevention, and treatment of VAP. Research should also include details of bundles used to reduce VAP incidence, protocols of care, and toolkits used to carry out root-cause analysis on VAP infections. Furthermore, it would be useful to have an overarching multi-Gulf center surveillance program to continually monitor the incidence and prevalence of VAP and HAP and, indeed, other HAIs across the GCC countries.