Prevalence and malignancy risk of incidental pulmonary nodules: insights from a retrospective study in Saudi Arabia

Abstract

Background

Incidental pulmonary nodules (IPNs) are frequently identified in computed tomography (CT) scans performed for unrelated clinical indications. While most are benign, some harbor malignancy, necessitating proper evaluation. This study assesses the prevalence, characteristics, and malignancy predictors of IPNs.

Methods

A retrospective review of CT scans conducted between January 2015 and December 2022 at a single institution was performed. From 16,433 CT reports, 749 cases of pulmonary nodules were identified, of which 219 met inclusion criteria. Of 219 incidental nodules identified, 103 (47%) had follow-up CT imaging and comprised the analytic cohort for outcome assessment. Demographic and clinical data, including smoking history and nodule characteristics, were analyzed. Malignancy was confirmed through histopathological examination. Statistical analyses included descriptive comparisons and penalized logistic regression to account for low event rates.

Results

The study involved 219 patients with an equal sex distribution. The median nodule count was 5 per patient, with a median size of 5 mm. Most nodules were in the right upper lobe. Malignancy was detected in 5 of 103 patients with follow-up imaging (4.85%), which is significantly higher compared with national registry data of the general population. While malignant nodules exhibited slightly larger median sizes (6.5 mm), they did not significantly differ from benign ones in terms of size distribution or lobar location (P>0.05). Patients above 50 years old showed a significantly higher number of nodules (P=0.002).

Conclusions

A significant proportion of IPNs in this cohort demonstrated malignancy risk, highlighting important gaps in follow-up and documentation in routine clinical practice. These findings highlight the need for structured, risk-based follow-up protocols to ensure timely evaluation while avoiding unnecessary interventions.

Highlight box.

Key findings

• Among 103 patients with follow-up imaging, malignancy was confirmed in 5 (4.85%), significantly higher than national population rates.

• Traditional clinical risk factors and most radiologic characteristics did not demonstrate statistically significant associations with malignancy.

• 47% of patients had documented follow-up imaging, highlighting major gaps in continuity of care.

What is known and what is new?

• Incidental pulmonary nodules (IPNs) are frequently detected on computed tomography imaging, and guideline-based follow-up is essential to balance early cancer detection with avoidance of unnecessary interventions.

• In this Saudi tertiary-care cohort outside formal screening programs, IPNs showed a relatively high malignancy yield despite limited predictive value of standard risk factors and poor follow-up adherence.

What is the implication, and what should change now?

• Implementation of structured, risk-based follow-up pathways and electronic nodule tracking systems is needed to improve evaluation and reduce missed malignancies in routine clinical practice.

Introduction

Incidental pulmonary nodules (IPNs) are a frequent finding in computed tomography (CT) scans, often identified during imaging conducted for unrelated clinical indications. These nodules, though predominantly benign, can occasionally represent early malignancies, necessitating careful evaluation to avoid missed opportunities for early cancer diagnosis while avoiding overdiagnosis (1). The increased detection rates of IPN generated using advanced imaging technology have caused concerns regarding the clinical treatment of these results (2-4).

The prevalence of incidental findings, especially pulmonary nodules, is noted across various settings. For instance, Barbosa et al. found pulmonary nodules to be the most common incidental findings, detected in approximately 45.7% of chest CTs performed in trauma patients (1). Wu et al. analyzed data from 413 patients with IPNs identified on abdominal CT and found that 11% were malignant^. (4). These findings emphasize the critical importance of early detection, as outlined by the Fleischner Society guidelines, which provide evidence-based recommendations for the management of pulmonary nodules detected on CT scans. These guidelines aim to stratify risk based on nodule size, appearance, and patient history to optimize follow-up and intervention strategies (5). However, despite the importance of early detection, adherence to follow-up recommendations for incidental findings remains suboptimal. Studies reveal that fewer than 14% of incidental findings receive timely follow-up, reflecting systemic challenges in ensuring adequate management of these patients (1,4).

Several factors influence the likelihood of malignancy in pulmonary nodules. Key predictors include nodule size, morphology, density, and growth patterns, as well as patient-specific variables such as age, smoking history, and comorbidities (1-4). For example, multivariate logistic regression models have demonstrated an 89% accuracy rate in identifying incidental findings on imaging scans, such as pulmonary nodules. This accuracy improves when incorporating key variables like smoking history, relevant medical diagnoses, and patient age, helping to differentiate between benign and potentially malignant nodules (1). Several multivariate logistic regressions–based models have been developed to estimate the probability of malignancy in pulmonary nodules, including the Mayo Clinic model, which incorporates age, smoking status, nodule size, spiculation and upper lobe location, and the Brock model, which additionally integrates nodule count, type and patient demographics (6,7). These models have demonstrated strong discriminatory performance and are widely used to guide risk‑based management.

Pulmonary cancer remains the leading cause of cancer-related mortality worldwide (3). While early-stage diagnosis significantly improves survival, approximately 70% of patients are diagnosed at advanced stages (2-4). Low-dose CT screening in randomized trials like The National Lung Screening Trial (NLST) and NELSON trial has shown substantial mortality reductions in high-risk populations, highlighting the significance of early detection and stratified management (2,3,5). This study aims to assess the prevalence, characteristics, and predictive factors for malignancy in IPNs detected on CT scans. This study aims to address the gap in the literature by enhancing our understanding of the outcomes of IPNs within our community. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2073/rc).

Methods

Study design and population

This retrospective study analyzed IPNs detected on chest CT scans performed between January 1, 2015, and December 31, 2022. Final data extraction occurred on March 15, 2023. The sample size was determined by all eligible cases during this study period. A formal power calculation was not conducted because this retrospective study analyzed all eligible cases within the study period. The sample size was predetermined by available data rather than recruitment. We therefore used penalized regression methods suitable for low‑event‑rate data to provide robust estimates. The study was conducted at King Saud University Medical City-King Khalid University Hospital, Riyadh, Saudi Arabia, a tertiary care center. The initial dataset included 16,433 patients who underwent chest CT scan either in inpatient or outpatient during the study period. A database query identified 749 patients with chest CT reports containing the terms “nodule” or “nodules”. Radiology and medical records underwent extensive review to refine the study population. Patients were excluded if they had a known medical history of cancer or were referred for metastatic assessments. Furthermore, we excluded patients who were symptomatic for pulmonary disease (e.g., presenting with cough, hemoptysis, or pleuritic chest pain) at the time of imaging and the symptoms were the reason the patient underwent the CT scan. This ensured that the pulmonary nodules identified were strictly incidental findings discovered during imaging for non-pulmonary indications. An IPN was defined as one discovered in asymptomatic patients during imaging conducted for reasons unrelated to pulmonary indications. A total of 219 patients with incidentally detected pulmonary nodules were identified during the study period. Among these, 103 patients had documented follow-up chest CT imaging and constituted the final analytic cohort for outcome and malignancy risk assessment. Patients without follow-up imaging (n=116) were excluded from outcome analyses. Patients were categorized into two groups: Group 1 (nonmalignant cases = benign stable nodules, resolved nodules, or non-malignant progressive nodules; n=98) and Group 2 (malignant nodules diagnosed as lung cancer; n=5) (Figure 1). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Ethical approval was obtained from the Institutional Review Board (IRB) of King Saud University, College of Medicine (No. E-23-7803) on 05 April 2023. Given the retrospective nature of the study and the use of anonymized data, the requirement for individual informed consent was waived by the IRB.

Figure 1.

Study flow diagram and analytic cohort selection. CT, computed tomography.

Data collection and variable

Demographic information, clinical characteristics, and smoking history were extracted from the electronic medical records of patients in the follow-up cohort. Imaging data, including nodule size, number, and location, were recorded for detailed analysis. All data on follow-up outcomes were collected to assess progression or resolution of the nodules. Reviewers were blinded to clinical outcomes during initial extraction. Data abstraction was cross-checked to reduce information bias. Malignancy was determined by histopathological confirmation for all positive cases. Benign nodules were defined as those demonstrating radiographic stability over 2 years or complete resolution during the follow-up period. Smoking status was defined as current or former smokers who had a lifetime exposure of at least 100 cigarettes, consistent with the Centers for Disease Control and Prevention U.S. (CDC) definition (8). Potential confounders included age, sex, smoking status, nodule size, family history of cancer, and comorbidities including coronary artery disease (CAD), hypertension, dyslipidemia, diabetes, and respiratory disease (including chronic obstructive pulmonary disease, asthma, interstitial lung disease, bronchiectasis, post-tuberculosis lung disease and pulmonary fibrosis). Variables were selected a priori based on published literature and clinical relevance to pulmonary nodule assessment.

Imaging protocols

This study focused exclusively on dedicated chest CT scans to ensure accurate and consistent assessment of pulmonary findings and to minimize selection bias. Whole-body CT scans and other imaging modalities were excluded because their primary purpose often differs, such as the whole-body CT scans usually used for evaluating metastatic disease in patients with known malignancies or assessing injuries in trauma settings. In such scenarios, nodules may be deprioritized or missed, given other clinical priorities. This could affect the true prevalence of IPNs. Additionally, the imaging protocols used in whole-body CT scans may lack the resolution and detail necessary for thorough pulmonary nodule evaluation. By including only dedicated chest CT scans performed with standardized protocols, this ensured consistent image quality and precise documentation of key parameters, including nodule size, number, and location. Data were extracted by two independent reviewers using a standardized template. A nodule was defined as any lesion measuring three centimeters or less, in accordance with the Fleischner Society Pulmonary Nodule Guidelines (5). Only solid nodules were included in the analysis; subsolid or groundglass nodules were not included because they require different surveillance protocols. Pulmonary nodules of all sizes, including micronodules measuring less than 6 mm, were included. Nodule definitions and size classification were standardized using the Fleischner Society Glossary of Terms for Thoracic Imaging (9). Nodule size was measured in millimeters on axial lung window images by an experienced thoracic radiologist using Picture Archiving and Communication System (PACS) measurement tools. Discrepancies were resolved by consensus through joint review by the two initial reviewers, with input from a senior thoracic radiologist.

Statistical analysis

Statistical analysis was conducted to compare characteristics between Group 1 (nonmalignant) and Group 2 (malignant). A two-sided P value <0.05 was considered statistically significant. Age was analyzed both continuously and dichotomized at 50 years, based on epidemiologic thresholds for lung cancer risk. Nodule size was analyzed in categories: ≤8 and >8 mm. The 8-mm cutoff is recognized by both the Fleischner Society and the American College of Chest Physicians (ACCP) as a threshold associated with higher malignancy risk (5,10). Firth regression was used due to low event rates, incomplete separation, and model non-convergence for both univariate and multivariate analysis. Patients without follow-up imaging were excluded from malignancy outcome analyses, and no imputation of missing data was performed, as this information was unavailable or incomplete. Sensitivity analysis was not required, as there were no missing data for variables included in the regression model. Data classified as missing not at random (MNAR) were excluded from the analysis due to a high rate of non-reporting. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). Statistical analyses were performed using SPSS Statistics version 25.

Results

Demographic and clinical characteristics

The study involved 219 patients with an almost equal sex distribution: 114 males and 105 females. At the initial CT scan, the average age was 57.39 years (males: 55.01 years, females: 59.90 years). Of the 219 patients identified with IPNs, 103 patients had available follow-up images. At the last follow-up, females remained slightly older (62.42 years) than males (57.09 years). The follow-up period ranged from a minimum of 3 months (0.25 years) to a maximum of 4.63 years, with a median follow-up of 1.62 years. Females generally had a longer follow-up duration (2.41 years) than males (1.92 years) (Table 1).

Table 1. Demographic and clinical characteristics of the follow-up cohort (n=103).

Variables	Males (n=52)	Females (n=51)	All
Age at last follow-up (years), mean	57.09	62.42	59.73
Follow-up duration (years), median (range)	1.92	2.41	1.62 (0.25–4.63)

Pulmonary nodule characteristics

The median number of nodules per patient was five, consistent across both sexes. Similarly, the median nodule size was 5 mm for all patients, showing no meaningful variation between males and females. When analyzing the distribution of nodules across lung lobes, the right upper lobe was most frequently involved (47.25%). Full lobe distribution is shown in Table 2.

Table 2. Pulmonary nodule characteristics by sex (n=219).

Variables	Males (n=114)	Females (n=105)	Overall (n=219)
Number of nodules per patient	5.00 [2]	5.00 [3]	5.00 [2]
Largest nodule size (mm)	5.00 [3]	5.00 [3]	5.00 [3]
Number right upper lobe	52 (45.61)	51 (49.04)	103 (47.25)
Number right middle lobe	9 (7.89)	8 (7.69)	17 (7.80)
Number right lower lobe	39 (34.21)	35 (33.65)	74 (33.94)
Number left upper lobe	41 (35.96)	46 (44.23)	87 (39.91)
Number left lower lobe	33 (28.95)	30 (28.85)	63 (28.90)

Data are presented as median [interquartile range] or n (%). Lobar distribution is reported as number of patients with nodules in that lobe; totals may exceed 100% because some patients had nodules in multiple lobes.

In univariate analysis, patients over 50 years of age had a significantly higher number of incidental nodules (P=0.002). However, age was not found to be a statistically significant independent predictor of malignancy in our specific cohort (OR =4.93, P=0.18).

Associated radiologic findings

Among the cohort, 37% had lymphadenopathy, with a slightly higher prevalence in females (39.05%) than in males (35.09%). Pleural effusion was identified in 8.22% of patients (males: 8.77%, females: 7.62%). Pleural thickening was less common, observed in just 4.17% of the cohort, slightly more in males (5.31%) than in females (2.91%) (Table 3).

Table 3. Associated radiologic findings in the baseline cohort (n=219).

Characteristics	Males (n=114)	Females (n=105)	Overall (n=219)
Lymphadenopathy	40 (35.09)	41 (39.05)	81 (37.00)
Effusion	10 (8.77)	8 (7.62)	18 (8.22)
Pleural thickening	6 (5.31)	3 (2.91)	9 (4.17)

Data are presented as n (%). Lymphadenopathy defined as mediastinal or hilar lymph node short axis ≥10 mm.

Risk factors

In the univariate analysis, all examined variables showed nonsignificant associations with malignancy (P>0.05). Positive associations were observed for CAD (OR =2.18), dyslipidemia (OR =2.52), family history of cancer (OR =3.46), hypertension (OR =1.79), and smoking (OR =2.27), while diabetes mellitus demonstrated a negative association (OR =0.89).

In the multivariate model, adjustment for covariates did not materially alter the direction or magnitude of associations. Positive but nonsignificant relationships persisted for CAD (adjusted OR =1.68, 95% CI: 0.14–12.28), dyslipidemia (adjusted OR =2.45, 95% CI: 0.35–18.18), family history of cancer (adjusted OR =3.16, 95% CI: 0.02–40.32), hypertension (adjusted OR =1.66, 95% CI: 0.21–13.76), and smoking (adjusted OR =1.75, 95% CI: 0.16–10.65), whereas diabetes mellitus remained negatively associated (adjusted OR =0.52, 95% CI: 0.07–3.45). None of the predictors reached statistical significance after adjustment (all P>0.05) (Table 4).

Table 4. Medical and historical risk factors for malignancy in pulmonary nodules.

Variables	Unadjusted			Adjusted
	OR (95% CI)	P value		OR (95% CI)	P value
Coronary artery disease	2.18 (0.21–12.35)	0.45		1.68 (0.14–12.28)	0.65
Diabetes mellitus	0.89 (0.15–4.69)	0.89		0.52 (0.07–3.45)	0.49
Dyslipidemia	2.52 (0.41–13.33)	0.29		2.45 (0.35–18.18)	0.35
Family history of cancer	3.46 (0.03–37.81)	0.49		3.16 (0.02–40.32)	0.53
Hypertension	1.79 (0.34–10.90)	0.49		1.66 (0.21–13.76)	0.63
Smoking	2.27 (0.22–12.89)	0.43		1.75 (0.16–10.65)	0.60

Firth penalized logistic regression. Outcome = malignancy (5/103). CI, confidence interval; OR, odds ratio.

Malignancy risks

In the univariate models, age emerged a positive but statistically non-significant independent predictor of malignancy in our specific cohort (OR =4.93, 95% CI: 0.55–650.14, P=0.18). Other radiologic findings, including adenopathy (OR =0.56, 95% CI: 0.06–3.08, P=0.52), effusion (OR =0.97, 95% CI: 0.01–9.08, P=0.98), and pleural thickening (OR =1.95, 95% CI: 0.01–19.42, P=0.69), were not significantly related to malignancy risk. Similarly, lesion localization across lobes (left middle, left upper, right lower, right middle, and right upper) and large nodule size showed no significant associations. In the multivariable Firth model adjusting for all predictors concurrently, none of the variables achieved statistical significance. Age remained positively associated with malignancy (adjusted OR =5.30, 95% CI: 0.54–744.94, P=0.19), and large nodule exhibited a similar directional effect (adjusted OR =5.04, 95% CI: 0.67–37.06, P=0.11). Other features, such as adenopathy (adjusted OR =1.04, 95% CI: 0.10–6.68, P=0.97) and pleural thickening (adjusted OR =1.83, 95% CI: 0.01–48.09, P=0.77), were not independently associated with malignancy. Overall, the penalized maximum likelihood model indicated no strong predictors, suggesting limited power or data sparsity across strata (Table 5).

Table 5. Associated radiologic findings in patients with incidental pulmonary nodules (n=103).

Variables	Unadjusted			Adjusted
	OR (95% CI)	P value		OR (95% CI)	P value
Adenopathy	0.56 (0.06–3.08)	0.52		1.04 (0.10–6.68)	0.97
Age >50 years	4.93 (0.55–650.14)	0.18		5.30 (0.54–744.94)	0.19
Effusion	0.97 (0.01–9.08)	0.98		0.49 (<0.01–7.04)	0.64
Large nodule (>8 mm)	3.87 (0.62–20.68)	0.13		5.04 (0.67–37.06)	0.11
Left middle lobe	0.22 (<0.01–1.95)	0.21		0.26 (<0.01–2.29)	0.27
Left upper lobe	2.15 (0.41–13.11)	0.36		3.07 (0.55–22.60)	0.20
Pleural thickening	1.95 (0.01–19.42)	0.69		1.83 (0.01–48.09)	0.77
Right lower lobe	0.17 (<0.01–1.53)	0.13		0.33 (<0.01–3.02)	0.37
Right middle lobe	1.02 (0.01–9.63)	0.99		2.34 (0.02–33.73)	0.62
Right upper lobe	1.58 (0.30–9.66)	0.58		1.16 (0.20–7.54)	0.86

Firth penalized multivariable logistic regression. CI, confidence interval; OR, odds ratio.

Lung cancer incidence

Among patients with available follow-up imaging (n=103), malignancy was detected in 5 patients (4.85%; 48.5 per 1,000). Using follow-up time to account for person-time at risk (~166.9 person-years based on median follow-up of 1.62 years), the estimated lung cancer incidence rate in the follow-up cohort was 2,996 per 100,000 person-years, compared with the national lung cancer incidence rate of 3.1 per 100,000 person-years. This comparison reflects risk enrichment in an incidental nodule cohort and is not age standardized (Table 6).

Table 6. Incidence of lung cancer in patients with incidental pulmonary nodules compared to the general population.

2020 cancer incidence report in Saudi Arabia	Incidental pulmonary nodule	P value
3.1 per 100,000 person-years	2,996 per 100,000 person-years	<0.001

Discussion

This study assessed the prevalence and malignancy predictors of IPNs in a Saudi Arabian cohort. Our primary results indicate a high incidence of nodules, with malignancy rates significantly exceeding general population averages. However, traditional risk factors and nodule characteristics did not yield statistically significant predictors, likely due to the limited sample size. Although the challenges of IPN follow-up and the relatively low malignancy rate are well described in the literature, most available data derive from Western populations or structured screening cohorts. This study provides real-world data from a Middle Eastern population outside formal lung cancer screening programs, reflecting routine clinical practice.

Nodules were predominantly located in the upper lobes, particularly the right upper lobe (47.25%). This distribution aligns with findings from Rinaldi et al., who reported that airflow patterns and carcinogen deposition contribute to upper lobe dominance (11). While some studies associate lower lobe nodules with malignancy due to proximity to basal vascular supply, which may facilitate metastatic seeding, this study’s findings did not support this trend (6). The size of nodules is a well-established predictor of malignancy, with larger nodules (>8 mm) showing higher malignant potential, as demonstrated by multiple studies (7,12). However, this study did not find significant differences in nodule size between malignant and non-malignant cases, diverging from the Fleischner Society guidelines, which stratify risk based on nodule size and appearance (5).

In line with American Cancer Society results that age is a major factor in nodule prevalence, the rise in nodule counts among patients over 50 years (P=0.002) indicates cumulative environmental and occupational exposures over time (13). However, the absence of a clear correlation between a higher number of nodules and cancer highlights the significance of personalized patient assessment and advanced imaging for risk assessment. Regarding traditional comorbidities such as hypertension and diabetes showed no significant association with malignancy in this cohort. This finding is consistent with other studies that reported that comorbidities are poor predictors of malignancy in incidental nodules (14,15). However, it could be a good predictor when combined with detailed smoking history which improves malignancy risk prediction in certain populations (16).

The reported lung cancer incidence in patients with incidental nodules is significantly higher than the general population rate in Saudi Arabia (17). This is expected given that the cohort consisted exclusively of patients with radiologically confirmed pulmonary nodules, and the comparison serves to emphasize the high stakes of under-followed incidental findings rather than to imply population prevalence. This aligns with international observations highlighting the need for close monitoring of incidental findings, even in populations traditionally considered low risk (5,17). These data support recommendations advocating structured follow-up protocols, including risk-adapted surveillance intervals and advanced imaging techniques (18). In our study, a substantial number of patients were excluded due to lack of follow-up, underscoring a critical gap in IPN management. Given the incidental nature of these findings, pulmonary nodules may be unintentionally deprioritized when imaging is performed for other clinical concerns, particularly in high-pressure clinical settings.

In this context, the reliance on manual review of CT reports and the absence of an electronic nodule tracking system represent important workflow limitations. Evidence indicates that artificial intelligence (AI) based applications can improve pulmonary nodule detection, standardize characterization, and enhance reporting consistency in routine CT imaging, thereby facilitating identification of nodules requiring follow-up (19). Integration of AI-assisted detection and structured reporting may therefore help address the workflow gaps observed in this cohort and support more consistent follow-up of IPNs.

Given the observed incidence, adoption of evidence-based guidelines would be instrumental in improving IPN management in Saudi Arabia. The Fleischner Society Guidelines provide a structured, risk-based framework for evaluation and follow-up of pulmonary nodules detected on CT, and adherence to these recommendations has been shown to limit unnecessary investigations while maintaining diagnostic safety and reducing healthcare costs (16). Implementing similar guideline-based pathways could promote consistency in clinical practice and reduce variability in decision-making. In parallel, lung cancer screening frameworks such as Lung-RADS standardize reporting and management of findings on low-dose CT screening (20). While screening eligibility was not assessed in this study, such frameworks illustrate how structured pathways can support early detection and organized follow-up. In this cohort, nodules were identified on standard-dose chest CT performed for routine indications; however, evidence indicates that low-dose CT protocols can achieve comparable diagnostic performance for nodule detection and longitudinal surveillance with substantially reduced radiation exposure making them a suitable option for follow-up imaging (21).

Limitations

This study has some limitations, including its retrospective, single-center design, which may introduce selection bias. Additionally, the exclusion of 116 patients due to the absence of follow-up imaging could contribute to attrition bias and impact the generalizability of the findings. Many patients did not have documented follow-up imaging, and the reasons were not always clear. Because our system does not track incidental nodules electronically, we were unable to confirm whether follow-up occurred outside the hospital or was simply not done. These workflow gaps likely contributed to the high loss-to-follow-up rate. The progress of slow-growing cancers might not be well captured by the comparatively brief median follow-up period (1.62 years). The statistical power of this study to detect significant relationships was constrained by the small number of confirmed malignant cases (n=5). As a retrospective study, the sample size and case distribution were dependent on available historical data and were not under investigator control. The lack of automated detection software for the initial screening of the 16,433 CT reports may have impacted the total number of nodules identified for inclusion, potentially underpowering the final analysis. The results may also be impacted by unmeasured environmental or occupational exposures. Furthermore, as the findings were incidental, the imaging reports primarily focused on nodule size and location, without detailed documentation of morphology or other characteristics. Detailed pack-year smoking data were inconsistently documented, which limited the ability to apply screening-based smoking thresholds. Restricting inclusion to dedicated chest CT examinations may limit generalizability, as IPNs can also be detected on imaging performed for other indications. However, this approach ensured consistent lung coverage and reliable nodule characterization, reducing heterogeneity. This limits the ability to assess features that could help differentiate between malignant and benign nodules.

Recommendations

Optimizing IPN management requires standardized follow-up protocols, advanced diagnostics, and multidisciplinary care to enhance malignancy risk assessment and decision-making. Future research should focus on multicenter studies and the integration of AI software for the automated detection and characterization of nodules. Implementing AI tools in future retrospective reviews could significantly increase the number of included patients and nodules, providing the necessary statistical power to identify subtle predictors of malignancy that were not reachable in this cohort (19). Longitudinal studies are needed to track nodule progression, while integrating biomarkers could improve malignancy risk stratification. Additionally, future research should also address the role of patient and physician education in the management of IPNs. Educating patients about the significance of follow-up and the potential risks of missed evaluations is crucial for improving adherence to recommendations. At the same time, targeted education for physicians, particularly primary care providers, is essential to ensure that incidental findings are appropriately communicated and managed according to evidence-based guidelines. These guidelines, specifically designed for incidentally found pulmonary nodules, help reduce variability in decision-making among healthcare providers and promote standardized, high-quality care. Additionally, initiating a national lung cancer screening program using low-dose CT scans for high-risk populations would enable early detection, facilitate care pathways, and improve outcomes by addressing current gaps in follow-up and timely intervention. To complement these national efforts, institutions could implement automated notifications and auto-referral options within radiology workflows. These system-level tools can help direct patients with actionable findings to the most appropriate care pathway, improving continuity of care.

Conclusions

A significant proportion of IPNs in this cohort demonstrated malignancy risk, highlighting the importance of standardized risk-based follow-up pathways for incidental findings detected in routine clinical practice. This study highlights existing gaps in follow-up and documentation of IPNs within a real-world healthcare setting outside formal screening programs. Our findings emphasize the need for structured follow-up protocols and guideline-based management to support timely detection of malignancy. Future multicenter and longitudinal studies are warranted to refine risk stratification and improve management pathways for IPNs.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Ethical approval was obtained from the Institutional Review Board of King Saud University, College of Medicine (No. E-23-7803). The requirement for individual informed consent was waived due to the retrospective design and anonymized data.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-aw-2073/dss

jtd-18-02-150-dss.pdf ^{(100.8KB, pdf)}

DOI: 10.21037/jtd-2025-aw-2073