Microbial Isolates and Antimicrobial Resistance Patterns in Adults with Inborn Errors of Immunity: A Retrospective Longitudinal Analysis of Sputum Cultures Free

Inborn errors of immunity (IEI) comprise a range of genetic disorders impacting the immune system. Those affected have an increased risk of developing various systemic conditions, including severe infections, malignancies, autoimmune diseases, and allergies [1]. IEIs exhibit distinct susceptibility patterns to infectious agents. For example, patients with humoral deficiencies are vulnerable to encapsulated bacteria, as antibodies primarily defend against extracellular pathogens, yet they can manage intracellular infections. In contrast, combined immunodeficiency exposes individuals to opportunistic pathogens like viruses, mycobacteria, protozoa, and fungi due to T-cell deficiencies. Chronic granulomatous disease (CGD) patients with impaired phagocyte oxidative burst function are at risk from mycobacteria, fungi, and specific bacteria (e.g., Staphylococcus aureus, Escherichia coli, Klebsiella pneumoniae, Burkholderia cepacia, Serratia marcescens, and Nocardia species).

Individuals with IEIs often receive broad-spectrum antibiotics, including prophylaxis. These antibiotics, combined with the specific type of IEI, influence infection patterns and antimicrobial susceptibility, which can change over time in a patient. The emergence of antimicrobial resistance in this context raises significant concerns, impacting both individual patient care and the broader population. There is a potential for resistant species to spread within the community, with studies indicating a 1.3- to 2-fold higher risk of infection-related mortality in patients infected with antibiotic-resistant bacteria [2]. Additionally, such infections result in longer hospital stays and costlier, broad-spectrum antibiotic treatments leading to increased healthcare expenditures [3].

Respiratory infections, which are both common and frequently the initial sign of IEI, are a leading cause of hospitalizations and fatalities in individuals with this condition; reducing infection burden is essential for life expectancy [4‒6]. Frequently, the causative pathogens are encapsulated bacteria, but viral infections also occur [7, 8]. Standard preventive measures include immunoglobulin replacement therapy (IgRT), facial masks and social distancing, and prophylactic antibiotics [9‒11]. Despite these measures, many IEI patients continue to suffer respiratory infections, which increases the risk of bronchiectasis [12‒14]. In a cohort of common variable immunodeficiency (CVID) patients followed over a mean of 11 years, 34.2% had chronic lung disease at diagnosis, and this figure increased to 46.3% during follow-up despite the use of IgRT [15].

Studies have shown that antimicrobial prophylaxis is effective in CGD, with patients often receiving prophylactic trimethoprim-sulfamethoxazole (TMP/SMX), which is effective against S. aureus, Nocardia, Burkholderia, and Serratia, and itraconazole for the prevention of fungal infection [16, 17]. The practice of prescribing antimicrobial prophylaxis has been extended to other IEI forms, with the patients receiving various agents, including TMP/SMX, macrolide, and other beta-lactam antibiotics. However, widespread use of prophylaxis carries the risk of promoting drug-resistant pathogens, potentially resulting in difficult-to-treat infections. Comprehensive microbiological analysis is crucial to identify causative agents and assess their susceptibility profiles, which allows for selecting appropriate antimicrobial agents for pathogen eradication [11]. Additionally, this approach can address important considerations, including cost, toxicity, and the choice of narrow-spectrum antibiotics to mitigate the development of drug-resistant isolates. Sputum culture is the primary microbiological test employed for identifying respiratory pathogens.

This study aimed to quantify the resistance patterns against commonly used oral and prophylactic antibiotics in patients diagnosed with IEI, a group necessitating regular prophylactic antibiotic use. Additionally, through analysis of this resistance data, our goal was to inform the strategic selection of antibiotics for empirical treatment regimens. This study investigated the causative agents responsible for respiratory infections among adult patients with IEIs. We have retrospectively analyzed the sputum culture results, with specific emphasis on the antimicrobial resistance patterns among different isolates and their correlation with a number of factors, including time, underlying etiology, receipt of antimicrobial prophylaxis, and the therapeutics employed to treat various manifestations. To our knowledge, this is the only report to investigate the characteristics of respiratory infections in IEI patients.

This retrospective comparative cohort involved a review of medical records pertaining to IEI patients with a history of positive sputum cultures. We systematically collected data on their clinical characteristics and demographics. The study protocol was approved by the Marmara University of Medical Science Ethics Committee (ID number 1353), and the patients were under the care of the Allergy Immunology clinic at a tertiary hospital in Istanbul.

Inclusion criteria for this study were as follows: (i) a confirmed diagnosis of IEI, (ii) age over 18 years, and (iii) a history of multiple positive sputum cultures. Patients with secondary immunodeficiency and those without positive sputum cultures were excluded from the study. Specific IEI diagnoses were made following the ESID criteria [18]. The classification of patients with IEI was made based on the International Union of Immunological Societies (IUIS). All patients had next-generation sequencing testing for underlying genetic defects (see online suppl. Table S1; Fig. S1; for all online suppl. material, see https://doi.org/10.1159/000541533).

Due to the nature of the study, healthy subjects were not suitable as controls. Consequently, individuals with chronic pulmonary disease such as bronchiectasis (non-cystic fibrosis [CF]) and primary ciliary dyskinesia (PCD) who experienced frequent infections were recruited as the control cohort. Patients diagnosed with PCD and non-CF bronchiectasis aged over 17 years and had positive sputum cultures were included.

Data collection spanned from 2014 to 2023, utilizing the electronic recording system. We meticulously documented information related to past sputum cultures, hospitalization history, the frequency of respiratory infections (particularly pneumonia), annual antibiotic prescriptions, and antibiotic sensitivity patterns for each bacterial strain. We considered sputum as “purulent” when >10 granulocytes per high-power field (hpf) were observed. Bacterial colonization was defined as the isolation of the same bacteria on more than two occasions, with at least a 3-month gap within the same year [19]. Multiple sputum samples from each patient were analyzed and plotted on a time axis to visualize longitudinal changes over the observation period.

All strains were identified using Matrix-Assisted Laser Desorption Ionization Time-of-Flight Mass Spectrometry (MALDI-TOF MS), supplemented by conventional tests for specific bacteria in our clinical microbiology laboratory. Antibiotic susceptibility tests for Streptococcus pneumoniae, Haemophilus influenzae, and Pseudomonas aeruginosa were conducted using the disk diffusion method on Mueller Hinton agar, with results interpreted according to EUCAST guidelines. Quality control was maintained using standard strains for each bacterium.

All data were analyzed with SPSS statical software package version 22 (IBM Corp., USA) and GraphPad Prism 8 (GraphPad Software Inc. San Diego, CA, USA). Median and interquartile range (IQR) values for continuous variables and the frequency and percentage for the categorical variables were calculated. Differences between ordinal data were evaluated with the Mann-Whitney U test and the Kruskal-Wallis test. Categorical variables were evaluated with the 2-tailed χ² or Fisher’s exact tests. Correlation tests were assessed with the Spearman correlation test. A p value <0.05 was considered the significance level for differences.

Ninety IEI patients underwent screening, and 36 had at least one positive sputum culture. A total of 395 sputum cultures were noted for those 36 patients between 2014 and 2023. Of the 395 cultures, 314 that yielded growth were included in the analysis. Eighty-one cultures showed normal flora and were therefore excluded from the evaluation (Fig. 1 shows the results of all sputum cultures of IEI patients). The median age of patients was 23.5 (IQR; 21–29.5) years, and 61% were male. Patients diagnosed with PCD (n = 34) and non-non-cystic fibrosis bronchiectasis (n = 20) were assessed as control groups. Tables 1 and 2 present an overview of the patients’ and control groups’ demographic and clinical features.

We documented 992 symptomatic respiratory infections requiring antibiotic prescribing in 36 patients during the study period, averaging 3.06 exacerbations per patient/year. There were 43 hospitalizations due to pneumonia in IEI patients.

Notably, 22 patients (61.1%) were on regular prophylactic antibiotics, with 13 (59%) receiving TMP/SMX prophylaxis. Within this group, 10 (77%) of 13 patients developed resistance to TMP/SMX. Antibiotic sensitivity testing of sputum cultures in patients receiving regular prophylactic antibiotics was compared with patients not receiving prophylactic antibiotics and presented in online supplementary Table S2. Antibiotic resistance patterns were similar between groups except for erythromycin and tobramycin, which had a higher resistance rate in prophylactic antibiotics groups (p: 0.04, p: 0.02, respectively).

As expected, the presence of bronchiectasis and the number of affected lobes were higher in sputum-positive patients. Additionally, the administration of IgRT was more frequently observed in the sputum-positive cohort. No significant differences were noted in other characteristics between the sputum-positive and negative groups (See online suppl. Table S3).

Out of the 314 positive sputum cultures, the most commonly detected bacteria was H. influenzae, followed by P. aeruginosa and S. pneumoniae in IEI patients (Table 2). H. influenzae was isolated in 159 (51%) cultures from 32 patients. P. aeruginosa, being the second most common organism, was isolated in 9 patients (a total of 51 isolates). Figure 2 shows all microbiologic isolates from sputum cultures of the patients with IEI. Similar to the IEI cohort, H. influenzae emerged as the most commonly isolated bacterium in the PCD cohort, while P. aeruginosa was the most prevalent organism in the non-CF bronchiectasis group (Table 2). Patients with IEI were assigned to predominantly antibody deficiency (PAD) and non-PAD to address the differences between diagnostic groups. H. influenzae and S. pneumoniae were detected more frequently in the PAD group (online suppl. Table S4).

In isolates of H. influenzae, resistance was observed against three classes of antibiotics: beta-lactams, sulfonamides, and quinolones. The antibiotic sensitivity patterns for the detected isolates are displayed in Table 2.

Among H. influenzae isolates in IEI patients, the rate of beta-lactamase positivity was 49%, and the most frequent resistance was toward cefuroxime (82%), followed by amoxicillin/clavulanic acid (66%). Also, the TMP/SMX resistance rate was 59%, and nearly half of H. influenzae positive cultures had resistance to ampicillin/sulbactam (49%). Among IEI patients, the resistance rate of quinolone was 24.2% for H. influenzae. TMP/SMX was the most frequently prescribed prophylactic antibiotic in the study group, and the drug resistance rate for this drug in H. influenzae isolates was 59%.

In IEI of patients, sensitivity testing for TMP/SMX in H. influenzae isolates unveiled the presence of three distinct subgroups: the first group remained sensitive (8 patients), the second one was consistently resistant (13 patients), and the third group showed a mixed pattern of sensitivity and resistance (11 patients) over time (see Fig. 3a). Notably, across all three subgroups, our analysis revealed that these patterns were not related to gender, age, serum immunoglobulin level, the presence or type of bronchiectasis, or the abnormalities in lymphocyte subsets. Among H. influenzae isolates in IEI patients, antibiotics sensitivity testing for other antibiotics than TMP/SMX was displayed in Figure 3b and c.

Considering that H. influenzae was the most frequently isolated bacterium in the majority of patients, the study focused exclusively on evaluating the resistance patterns of H. influenzae in these patients over the years. In an analysis of 16 patients with 3 or more positive H. influenzae cultures, resistance to amoxicillin-clavulanic acid emerged in 9 (60%) patients over time. Two patients continued to show sensitivity in their latest cultures, while 4 (26.6%) patients exhibited consistent resistance across all cultures. Ampicillin-sulbactam resistance developed in 8 (50%) patients, with 6 (37.5%) remaining sensitive in their latest cultures and 2 showing consistent resistance across all cultures. For cefuroxime, 7 (43.7%) patients experienced a shift from sensitivity to resistance, 6 (37.5) demonstrated persistent resistance, and 3 remained sensitive in the recent cultures. Ceftriaxone resistance was observed in only 1 patient, with 14 (93.3%) patients showing sensitivity (see Fig. 3b, c).

In the control groups, which did not receive regular prophylactic antibiotics, the TMP/SMX resistance rate in H. influenzae isolates was significantly lower than in the study group (59% vs. 11% and 6%, p < 0.001). Additionally, the resistance rate for cefuroxime and ampicillin/sulbactam was significantly higher in IEI patients than in control groups (p < 0.001 and 0.0008, respectively). Table 2 provides a detailed comparison of the results of antibiotic sensitivity testing between the groups.

In isolates of P. aeruginosa, resistance was noted against three groups of antibiotics: beta-lactams, quinolones, and aminoglycosides. Among the P. aeruginosa isolates in IEI patients, the most frequent drug resistance was with ciprofloxacin (85%). The ceftazidime resistance rate was 42%, and aminoglycoside antibiotic resistance ranged from 23% to 33%. Among IEI patients, the carbapenem resistance rate was 29.6%, while the resistance rate of quinolone (ciprofloxacin) was 48.1%. Figure 4 shows antibiotic sensitivity testing in all P. aeruginosa isolates in the study group. The results of antibiotic sensitivity testing revealed a higher rate of resistance among IEI patients compared to control groups for Pseudomonas aeruginosa isolates (Table 2).

In S. pneumoniae isolates, resistance was identified against six groups of antibiotics: beta-lactams, sulfonamides, quinolones, tetracyclines, macrolides, and lincosamides. Within the S. pneumoniae isolates, all samples in IEI patients exhibited resistance to tetracycline (17/17, 100%). Moreover, the resistance rates for erythromycin, clindamycin, and penicillin were notably high, at 89.6%, 92.3%, and 95.6%, respectively. The resistance rates for penicillin, ceftriaxone, and TMP/SMX were observed to be higher in IEI patients compared to those in PCD patients, with significant differences noted (p = 0.003, p = 0.004, and p = 0.01, respectively) (see Table 2). Figure 5 provides a comprehensive visual representation of S. pneumoniae isolates detected over time and the results of sensitivity testing for all IEI patients.

The research revealed that among 12 IEI patients, sputum cultures showed colonization by H. influenzae in all 12 cases. Additionally, 3 of these patients were concurrently colonized with S. pneumoniae and P. aeruginosa. The rate of bacterial colonization was similar between IEI and PCD groups (Table 2). Furthermore, bronchiectasis was detected in 31 patients (31/36, 86.1%), with a higher prevalence of tubular bronchiectasis than the cystic type.

Throughout the follow-up period, a total of 38 co-infections were documented in 13 IEI patients. The predominant co-infection was the concurrent presence of H. influenzae and S. pneumoniae. Furthermore, diverse combinations of pathogens, including P. aeruginosa, Moraxella catarrhalis, S. pneumoniae, E. coli, S. aureus, Serratia spp., and Providencia spp., were detected in 25 instances of co-infections. Only 1 patient with Bruton disease had multidrug-resistant tuberculosis in the 9 years following.

According to the findings of this study, the most commonly isolated bacteria from the sputum of adult IEI patients were H. influenzae, P. aeruginosa, and S. pneumoniae. Notably, antibiotics frequently prescribed for these patients, including TMP/SMX, cephalosporin, tetracycline, penicillin, erythromycin, clindamycin, and ciprofloxacin, exhibited significant resistance rates. Patients with non-CF bronchiectasis and PCD had similar bacterial isolates in sputum cultures. However, the resistance rate to TMP/SMX, cephalosporins, and quinolones was lower than in patients with IEI.

While it is generally accepted that IgRT and prophylactic antibiotics should be considered for IEI patients with poor antibody responses and frequent infections, many patients continue to experience severe and recurrent infections. It is important to mention that recommendations regarding antibiotic prophylaxis are primarily based on anecdotal observations and expert opinions, often stemming from experiences with specific conditions like CGD, cystic fibrosis, or non-CF bronchiectasis. Conducting placebo-controlled studies to evaluate the efficacy and safety of antibiotic prophylaxis in IEI patients is challenging due to the severity of these diseases and the potential risks involved.

In addition to prophylaxis, IEI patients often require frequent and prolonged antibiotic treatments, which can contribute to the development of antibiotic resistance. This resistance can lead to treatment failure, long-term organ complications, and even fatalities. Given the emergence of antimicrobial resistance, there is a growing healthcare burden associated with the use of broad-spectrum antibiotics. To address this issue, some experts recommend periodically switching between different antibiotics, typically every 6 months, to mitigate the risk of developing antibiotic resistance.

TMP/SMX prophylaxis has been widely used in CGD patients, with favorable outcomes supported by extensive experience. This prophylactic approach extends beyond CGD and is commonly employed for other IEI patients as well. Our current study observed a notably high resistance rate to TMP-SMX (59%) among H. influenzae isolates. Unfortunately, there is a lack of comprehensive data on antibiotic resistance rates among patients receiving prophylactic antibiotics in the existing literature. We found that patients with IEI exhibited higher resistance rates to TMP/SMX in sputum isolates compared to those observed in control groups. Interestingly, no significant difference in resistance rates was identified between patients undergoing prophylactic treatment and their untreated counterparts.

At our center, the decision to prescribe prophylactic antibiotics takes into account the presence of bronchiectasis and the patient’s history of drug-resistant isolates in previous sputum cultures. In cases where bronchiectasis is present, our preferred prophylactic antibiotics include macrolides, such as azithromycin. This strategy aims to not only reduce colonization by Pseudomonas species but also take advantage of the anti-inflammatory properties associated with these agents.

Within our hospital’s cohort, among patients without IEI, 12.3% of H. influenzae isolates exhibited beta-lactamase positivity. Additionally, resistance rates to quinolones and cephalosporins were observed at 3.3% and 26.2%, respectively, in these isolates. The rate of beta-lactamase positivity in H. influenzae isolates among IEI patients rose by 49%. Moreover, among IEI patients, notably higher resistance rates were observed: 82% for cefuroxime, 66% for amoxicillin/clavulanic acid, and 59% for TMP/SMX. On the other hand, within the PCD cohort, the rate of beta-lactamase positivity reached 27.7%, and the resistance rates to quinolones and cephalosporins were 2.7% and 16.6%, respectively. These figures highlight the growing challenge of outpatient treatment for H. influenzae infections.

A study on CVID patients reported that patients receiving prophylactic antibiotics had more frequent symptomatic exacerbation rates than patients not using prophylaxis (mean, 2.87 ± 2.21 vs. 1.71 ± 1.33; p = 0.02) [20]. Interestingly, among patients on prophylactic antibiotics, the higher number of symptomatic exacerbations did not differ in the presence of bronchiectasis. In that study, 56% of exacerbation was associated with viral pathogens; the most common viral agent was rhinovirus spp. Bacteria were isolated in 33% of the exacerbations, and the most common bacteria were H. influenzae, S. pneumoniae, and P. aeruginosa. Viral and bacterial co-infection was detected in 25% of exacerbations, and no pathogen was isolated in 27.5%. Receiving prophylactic antibiotics was found to be a risk factor for delays in commencing antibiotics for an exacerbation. The most frequent bacterial isolates in our study were H. influenzae, P. aeruginosa, and S. pneumoniae. Although 992 infection episodes were treated with antibiotics, bacteria were isolated in only 31.6%, indicating viral infections may play a role in many situations.

In a previous study on X-linked agammaglobulinemia (X-LA) patients, 34% of lung infections resulted from H. influenzae and S. pneumoniae while on IgRT. In the same study, the bronchiectasis rate was 39% in X-LA patients [21]. Also, S. pneumoniae, H. influenzae, Mycoplasma pneumonia, Pseudomonas spp., Staphylococcus spp., and Klebsiella pneumonia are the causative pathogens in patients with CVID, respectively [8]. A prospective study showed that patients with PAD exhibited more common respiratory tract infections, especially with upper respiratory viruses, despite receiving IgRT than healthy controls [22]. In another report with IgG subgroup deficiency and specific polysaccharide antibody deficiency in adult patients, there was a significant reduction in the number of infections, antibiotic usage, and hospital admissions after commencing IgRT [23].

Among S. pneumoniae isolates, tetracyclin, penicillin, erythromycin, and clindamycin showed the highest resistance rates. These findings suggest that challenging the oral treatment of S. pneumoniae infection is complicated. Quinolones could be considered for oral treatment of S. pneumoniae infections, with levofloxacin showing resistance in only 1 out of 11 isolates and moxifloxacin having a resistance rate of 11.6%.

The result of a study with pediatric IEI patients reported that antibiotic susceptibility testing in Pseudomonas isolates was highly sensitive to ciprofloxacin (68%) and imipenem (82%) [24]. However, in the same research, there was a lower sensitivity rate to aminoglycosides among the Pseudomonas isolates. Our results showed higher rates of ciprofloxacin resistance in Pseudomonas isolates and less remarkable resistance rates with the aminoglycosides. However, using ciprofloxacin in the pediatric population is challenging due to side effects. Therefore, higher ciprofloxacin sensitivity may be secondary to the reluctance among physicians toward prescribing this medication in the pediatric population. Also, our study population encompassed adult patients who have used antibiotics for longer durations than the pediatric population, a factor important in the evolution of drug-resistant isolates. Notably, the imipenem susceptibility rate was 28%, significantly lower than the pediatric group. With a carbapenem resistance rate of 28% observed among IEI patients, this class of antibiotics may be regarded as a viable empirical treatment option for managing Pseudomonas infections.

According to reports in the United Kingdom, 106 courses of antibiotics were prescribed for 1,000 men and 155 per 1,000 women for respiratory tract infections [25]. During the study period, 992 symptomatic respiratory exacerbations requiring antibiotic prescribing (averaging 3.06 per patient/year) were documented. Despite IgRT and optimal treatment, this heightened frequency of antibiotic prescriptions may be associated with the emergence of antimicrobial resistance and bacterial colonization.

This study has several limitations. Firstly, its retrospective nature poses risks for data availability, especially for patients treated as outpatients, as medical data recording tends to be less rigorous in such cases compared to inpatient care. However, our electronic health recording system provided comprehensive data on antibiotic use and other laboratory assessments. Secondly, our study exclusively focused on bacterial culture results, potentially overlooking infections of viral origin or cases with concomitant viral and bacterial infections that led to antibiotic prescriptions. Another limitation stemming from the study’s retrospective nature is the constrained availability of data on the serotypes of H. influenzae and S. pneumoniae.

Despite these limitations, our study sheds light on the significant challenges posed by emerging antibiotic resistance patterns among IEI patients, who frequently experience infections and are already prescribed various antibiotics. While our findings provide a general overview, it is crucial to document profiles from other centers and test the generalization ability of our results with larger datasets.

Our study indicates a notably higher TMP/SMX resistance rate in H. influenzae isolates among IEI patients compared to patients with PCD and bronchiectasis. Within the constraints of this study, it was not feasible to definitively ascertain the impact of TMP/SMX prophylaxis on resistance development among IEI patients. This area warrants further exploration in future research endeavors. Notably, S. pneumoniae isolates exhibited elevated resistance to commonly prescribed oral agents, including tetracycline, clindamycin, and erythromycin, positioning moxifloxacin as a viable alternative therapeutic option. Additionally, the prevalent resistance of P. aeruginosa isolates to ciprofloxacin underscores the complexities inherent in the outpatient management of Pseudomonas infections. This investigation further underscores the critical role of routine sputum culture collection, extending to asymptomatic intervals, in the comprehensive management of these infections. These findings underscore the necessity for larger sample-size trials focusing on antibiotic sensitivity in IEI patients. Such studies are crucial to establishing evidence-based approaches for both prophylactic and therapeutic management of these patients.

The authors thank Elif Karakoç Aydıner and Safa Barıs for their invaluable guidance, support, and expertise throughout the research process. Additionally, we acknowledge the assistance of artificial intelligence in improving our English language usage during the manuscript preparation.

The study was approved by the Ethics Committee of Marmara University, Faculty of Medicine (ID No: 1353). All patients provided written informed consent to collect and analyze their medical records and participate in the study. This study was conducted according to the Good Clinical Practice guidelines, the Declaration of Helsinki, and the local ethical and legal guidance.

The authors have no conflicts of interest to declare.

The authors received no funding for this manuscript.

E.K. designed the study, collected patients’ samples, and wrote the manuscript; A.O., intellectually contributed to the research and wrote the manuscript. A.I. collected data from the results of sputum cultures and evaluated the resistance rates. Y.G., H.M.V., and S.O.Y. gathered samples from control group patients and analyzed the results.

Edited by: H.-U. Simon, Bern.

All data generated and analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.