The Feasibility of Uniportal Video-Assisted Thoracic Surgery in Octogenarians: A Propensity-Matched Comparative Analysis

Abstract

Objectives: To evaluate the short-term safety (30-day and in-hospital morbidity and mortality) and technical feasibility of uniportal video-assisted thoracic surgery (U-VATS) for anatomical lung resection in octogenarians (≥80 years) compared with younger patients (<80 years) at a single center. Methods: Ninety consecutive patients undergoing U-VATS anatomical lung resections between January 2020 and January 2024 were retrospectively analyzed. Patients were stratified by age: 60 patients < 80 years and 30 octogenarians ≥ 80 years. Propensity score matching (nearest-neighbor, 1:2 ratio, caliper 0.2 SD) yielded a matched cohort of 60 patients (40 younger, 20 octogenarians) for comparative analysis. Results: After matching, standardized mean differences (SMD) were <0.25 for most covariates, indicating good balance. Octogenarians demonstrated lower FEV1 (75.2 ± 15.3% vs. 87.5 ± 18.2%, p = 0.012) and DLCO (68.4 ± 12.1% vs. 78.5 ± 14.3%, p = 0.009), consistent with age-related pulmonary changes. Charlson Comorbidity Index was higher (5.3 ± 1.2 vs. 3.8 ± 1.4, p = 0.001). Surgical parameters were comparable: operative time (143.80 ± 42.3 vs. 136.55 ± 38.7 min, p = 0.524), blood loss (median 80 [IQR 50–120] vs. 95 [IQR 60–130] mL, p = 0.742). Zero conversions occurred. Major complications (Clavien–Dindo ≥ 3) occurred in 10% vs. 0% (absolute risk difference 10%, 95% CI: −3.2% to 23.2%). No 30-day mortality. 90-day mortality: 5% vs. 0% (p = 0.333); one-year: 15% vs. 0% (p = 0.035). Conclusions: U-VATS is technically feasible in carefully selected octogenarians with comparable intraoperative parameters to younger patients. Postoperative recovery differed meaningfully, with higher delirium rates, longer hospitalization, and greater rehabilitation needs. One-year mortality was higher in octogenarians, reflecting competing comorbid risk rather than surgical harm. Residual imbalance in comorbidity burden and pulmonary reserve after matching limits causal inference. These hypothesis-generating findings support U-VATS in selected octogenarians when comprehensive geriatric assessment and structured delirium prevention guide perioperative management; validation in larger multicenter prospective studies is required.

1. Introduction

The global demographic transition presents unprecedented challenges in thoracic surgery. By 2050, the population aged 80 and above is projected to triple, reaching 426 million worldwide [1]. This demographic shift coincides with lung cancer’s persistence as the leading cause of cancer-related mortality, with peak incidence in the seventh and eighth decades [2].

Historically, advanced age has been considered a relative contraindication to major pulmonary resection. Traditional open thoracotomy in elderly patients has been associated with complication rates exceeding 40% and mortality rates of 5–10% [3,4]. However, minimally invasive techniques, particularly video-assisted thoracic surgery (VATS), have challenged these paradigms.

Recent evidence demonstrates that VATS significantly reduces postoperative complications, shortens hospital stays, and accelerates functional recovery compared to open approaches [5]. The evolution to uniportal VATS (U-VATS) represents further refinement, potentially offering additional benefits through reduced intercostal nerve trauma [6].

Despite technical advances, the application of U-VATS in octogenarians remains controversial. While several studies have demonstrated feasibility in elderly populations, most have focused on septuagenarians or combined elderly cohorts without specific analysis of the extreme elderly [7,8]. Furthermore, existing literature predominantly originates from high-volume centers with comprehensive multidisciplinary teams and alternative treatment modalities.

This study addresses this knowledge gap by evaluating the safety and feasibility of U-VATS anatomical lung resections specifically in octogenarians, using propensity score matching to minimize selection bias, in a single center where surgical resection remains the primary curative option.

Additionally, the previous literature has documented postoperative delirium rates of 17–25% in elderly patients undergoing thoracic procedures using validated assessment tools such as CAM-ICU and DeltaScan, underscoring the clinical significance of this complication [9,10]. Likewise, one-year non-cancer-related mortality in octogenarians after lung resection has been reported to reach 14–20%, emphasizing the importance of broader postoperative risk appraisal in this group [9,10,11].

2. Methods

2.1. Study Design and Patient Selection

This retrospective cohort study analyzed all consecutive patients undergoing U-VATS anatomical lung resections at The Edith Wolfson Medical Center between January 2020 and 22 January 2024. U-VATS anatomical resections were introduced at our institution in 2020; this timeframe was therefore selected to capture the complete institutional experience with this technique from program inception, ensuring consecutive case inclusion without truncation bias. The study received institutional review board approval (WOMC-0024-24), and informed consent was waived due to its retrospective nature.

Inclusion criteria comprised: (1) anatomical lung resection (lobectomy or segmentectomy) via U-VATS approach; (2) complete clinical and follow-up data; (3) minimum one-year follow-up or death within the follow-up period. Exclusion criteria included: (1) non-anatomical wedge resections; (2) conversion to multiportal VATS or thoracotomy for technical reasons unrelated to patient factors; (3) incomplete pathological data.

Patients were stratified into two groups: sexagenarians and septuagenarians (<80 years) and octogenarians (≥80 years). All procedures were performed by a single experienced thoracic surgeon (FAA) with over 200 U-VATS procedures.

2.2. Preoperative Assessment

Comprehensive preoperative evaluation included:

• Pulmonary function tests with FEV1 and DLCO (percentage predicted and absolute values).
• Cardiac assessment with echocardiography and stress testing when indicated.
• Charlson Comorbidity Index calculation.
• Nutritional assessment including albumin levels.
• Comprehensive geriatric assessment for octogenarians when feasible When performed, the assessment included components such as Mini-Cog screening, Barthel Index, Charlson Comorbidity Index, and medication review; it was conducted in approximately 70% of octogenarian patients.
• Anesthesia risk stratification using ASA classification.

Surgical candidacy was determined through multidisciplinary team discussion. The decision between lobectomy and segmentectomy was based on: (1) tumor size and location; (2) pulmonary reserve; (3) ability to achieve adequate margins (minimum 2 cm or tumor diameter, whichever greater).

2.3. Surgical Technique

All procedures utilized standard U-VATS technique:

• Single 3–4 cm incision at 5th intercostal space, anterior axillary line.
• Complete individual hilar dissection.
• Systematic lymph node dissection (minimum 3 N2 stations plus hilar/intrapulmonary nodes).
• Inflation-deflation technique for intersegmental plane identification in segmentectomies.
• Frozen section analysis for margins in segmentectomy cases.
• Fissureless technique when appropriate.

2.4. Perioperative Management

Enhanced Recovery After Surgery (ERAS) protocol was implemented with age-specific modifications:

• Multimodal analgesia with reduced opioid dosing (reported as morphine milligram equivalents [MME]).
• Early mobilization within 6 h when feasible.
• Chest physiotherapy three times daily.
• Delirium prevention using Confusion Assessment Method (CAM) monitoring: CAM was administered by trained nursing staff every 8 h beginning on postoperative day 0 and continuing until discharge. Positive CAM screens were confirmed by geriatric medicine consultation within 24 h. The institutional delirium prevention protocol included early mobilization, sleep hygiene optimization, sensory aid provision, and avoidance of deliriogenic medications where feasible.
• Nutritional optimization.
• Prophylactic anticoagulation per institutional protocol.

2.5. Propensity Score Matching

Propensity score matching is a statistical technique that reduces selection bias in observational studies by pairing each patient from one group with a comparable patient from the other group based on their probability of group membership, given a defined set of baseline covariates; this approach mimics, within the constraints of available data, the covariate balance that randomization would achieve in a controlled trial.

To minimize selection bias, propensity score matching was performed using the following covariates:

• Age;
• Sex;
• Body mass index (BMI);
• ASA score;
• Major comorbidities (COPD, ischemic heart disease).

Matching was performed using nearest neighbor method without replacement, with a 1:2 ratio (octogenarians: younger patients) and caliper width of 0.2 standard deviations of the logit of the propensity score. Balance was assessed using standardized mean differences (SMD), with SMD < 0.25 indicating acceptable balance.

Nearest-neighbor matching without replacement was selected for its simplicity, interpretability, and established performance in small-to-moderate sample sizes; it minimizes the distance between matched pairs while avoiding duplicate use of controls, which is particularly important given our limited octogenarian cohort.

2.6. Outcome Measures

Intraoperative outcomes:

• Operative time;
• Estimated blood loss;
• Conversion rate to thoracotomy.

Postoperative outcomes:

• 30-day and 90-day mortality (all-cause);
• One-year mortality (all-cause);
• Postoperative complications classified by Clavien–Dindo grade;
• Length of hospital stay;
• ICU admission rate and duration;
• Discharge disposition;
• Delirium incidence (CAM criteria).

2.7. Statistical Analysis

Statistical analysis was performed using SPSS version 28.0 and R version 4.1.0 (MatchIt package). Normality was assessed using Shapiro–Wilk test. Normally distributed continuous variables were expressed as mean ± standard deviation and compared using independent t-tests. Non-normally distributed variables were expressed as median (interquartile range) and compared using Mann–Whitney U tests. Categorical variables were compared using Fisher’s exact test. Absolute risk differences with 95% confidence intervals were calculated for key binary outcomes. Statistical significance was set at p < 0.05. All tests were two-tailed.

3. Results

3.1. Study Population and Propensity Score Matching

Of 90 consecutive patients who met inclusion criteria, 60 were sexagenarians/septuagenarians and 30 were octogenarians. Propensity score matching using a 1:2 ratio yielded a well-balanced cohort of 60 patients: 40 younger patients matched to 20 octogenarians. The matching process successfully reduced standardized mean differences (SMD) for key covariates, with most achieving SMD values below 0.25, indicating acceptable balance between groups (

Table 1. Baseline Characteristics Before and After Propensity Score Matching. (a): Pre-Matching Characteristics (n = 90); (b): Post-Matching Characteristics (n = 60).

(a)
Variable	<80 Years (n = 60)	≥80 Years (n = 30)	p-Value	SMD
Age, years	68.5 (64–73)	82.0 (81–84)	<0.001	3.542
Male sex, n (%)	38 (63.3)	17 (56.7)	0.544	0.135
BMI, kg/m2	26.8 ± 4.5	25.4 ± 3.8	0.151	0.328
Smoking history, n (%)	48 (80.0)	26 (86.7)	0.432	0.183
Pack-years	35 (20–50)	40 (25–55)	0.298	0.241
Charlson Comorbidity Index	3.5 ± 1.6	5.8 ± 1.3	<0.001	1.583
ASA score ≥ 3, n (%)	32 (53.3)	25 (83.3)	0.005	0.675
COPD, n (%)	15 (25.0)	12 (40.0)	0.141	0.324
Ischemic heart disease, n (%)	18 (30.0)	14 (46.7)	0.118	0.347
FEV1, % predicted	92.3 ± 19.8	72.8 ± 17.2	<0.001	1.049
DLCO, % predicted	81.2 ± 16.1	65.3 ± 14.5	<0.001	1.040
(b)
Variable	<80 years (n = 40)	≥80 years (n = 20)	p-value	SMD
Age, years	70.2 (66–74)	82.0 (81–84)	<0.001	2.956
Male sex, n (%)	24 (60.0)	11 (55.0)	0.715	0.101
BMI, kg/m2	26.3 ± 4.2	25.8 ± 3.9	0.658	0.123
Smoking history, n (%)	32 (80.0)	17 (85.0)	0.640	0.132
Pack-years	35 (20–50)	40 (25–55)	0.421	0.186
Charlson Comorbidity Index	3.8 ± 1.4	5.3 ± 1.2	<0.001	1.152
ASA score ≥ 3, n (%)	24 (60.0)	16 (80.0)	0.124	0.442
COPD, n (%)	12 (30.0)	8 (40.0)	0.445	0.211
Ischemic heart disease, n (%)	14 (35.0)	9 (45.0)	0.460	0.206
FEV1, % predicted	87.5 ± 18.2	75.2 ± 15.3	0.012	0.726
FEV1, L	2.34 ± 0.52	1.89 ± 0.48	0.002	0.902
DLCO, % predicted	78.5 ± 14.3	68.4 ± 12.1	0.009	0.760
DLCO, mL/min/mmHg	18.2 ± 4.3	14.6 ± 3.9	0.003	0.871
Albumin, g/dL	3.9 ± 0.4	3.8 ± 0.3	0.318	0.282

).

3.2. Baseline Characteristics of Matched Cohort

Following propensity score matching, the groups remained significantly different only in age-related parameters (Table 1b). The mean age was 70.2 years (IQR 66-74) in the younger group versus 82.0 years [IQR 81-84] in octogenarians (p < 0.001). Despite matching, octogenarians demonstrated significantly reduced pulmonary function with mean FEV1 of 75.2 ± 15.3% predicted compared to 87.5 ± 18.2% in younger patients (p = 0.012), and mean DLCO of 68.4 ± 12.1% versus 78.5 ± 14.3% predicted (p = 0.009). The Charlson Comorbidity Index remained higher in octogenarians (5.3 ± 1.2 vs. 3.8 ± 1.4, p < 0.001), while other matched variables including sex distribution (55.0% vs. 60.0% male, p = 0.715), BMI (25.8 ± 3.9 vs. 26.3 ± 4.2 kg/m², p = 0.658), and prevalence of COPD (40.0% vs. 30.0%, p = 0.445) showed no significant differences (Table 1b).

3.3. Surgical and Pathological Characteristics

The distribution of surgical procedures was similar between groups, with lobectomy performed in 75.0% of octogenarians versus 70.0% of younger patients (p = 0.690), and segmentectomy in the remaining cases (

Table 2. Surgical and Pathological Characteristics (Post-Matching).

Variable	<80 Years (n = 40)	≥80 Years (n = 20)	p-Value
Procedure type
- Lobectomy, n (%)	28 (70.0)	15 (75.0)	0.690
- Segmentectomy, n (%)	12 (30.0)	5 (25.0)	0.690
Lobe resected			0.854
- Right upper	12 (30.0)	7 (35.0)
- Right middle	2 (5.0)	1 (5.0)
- Right lower	8 (20.0)	3 (15.0)
- Left upper	10 (25.0)	6 (30.0)
- Left lower	8 (20.0)	3 (15.0)
Clinical stage			0.786
- IA	18 (45.0)	8 (40.0)
- IB	14 (35.0)	7 (35.0)
- IIA	6 (15.0)	4 (20.0)
- IIB	2 (5.0)	1 (5.0)
Histology			0.931
- Adenocarcinoma	24 (60.0)	13 (65.0)
- Squamous cell	12 (30.0)	5 (25.0)
- Other	4 (10.0)	2 (10.0)
Tumor size, cm	2.8 ± 1.2	2.9 ± 1.1	0.757
Pathological stage			0.865
- IA	16 (40.0)	7 (35.0)
- IB	12 (30.0)	6 (30.0)
- IIA	8 (20.0)	5 (25.0)
- IIB	2 (5.0)	1 (5.0)
- IIIA	2 (5.0)	1 (5.0)
Lymph nodes examined	18 (14–22)	16 (13–20)	0.432
R0 resection, n (%)	40 (100)	20 (100)	1.000

). Tumor characteristics were comparable, including mean tumor size (2.9 ± 1.1 vs. 2.8 ± 1.2 cm, p = 0.757) and histological distribution, with adenocarcinoma representing the majority in both groups (65.0% vs. 60.0%, p = 0.931). All patients achieved R0 resection with negative margins. The median number of lymph nodes examined was 16 (IQR 13-20) in octogenarians versus 18 (IQR 14-22) in younger patients (p = 0.432), meeting oncological standards for both groups (Table 2).

3.4. Intraoperative Outcomes

Operative parameters demonstrated no significant differences between groups (

Table 3. Intraoperative Characteristics (Post-Matching).

Variable	<80 Years (n = 40)	≥80 Years (n = 20)	p-Value
Operative time, minutes	136.55 ± 38.7	143.80 ± 42.3	0.524
Blood loss, mL	95 (60–130)	80 (50–120)	0.742
Conversion to thoracotomy, n (%)	0 (0)	0 (0)	1.000
Intraoperative complications, n (%)	1 (2.5)	1 (5.0)	0.999
Chest tubes placed, n	1 (1–1)	1 (1–1)	0.847
Use of energy devices, n (%)	38 (95.0)	19 (95.0)	1.000
Adhesiolysis required, n (%)	14 (35.0)	9 (45.0)	0.460

). Mean operative time was 143.80 ± 42.3 min in octogenarians compared to 136.55 ± 38.7 min in younger patients (p = 0.524). Median estimated blood loss was lower in octogenarians at 80 mL (IQR 50-120) versus 95 mL (IQR 60-130) in the younger group (p = 0.742). Notably, no conversions to thoracotomy occurred in either group. Adhesiolysis was required in 45.0% of octogenarians and 35.0% of younger patients (p = 0.460), while intraoperative complications were rare, occurring in one patient per group (5.0% vs. 2.5%, p = 0.999) (Table 3).

3.5. Postoperative Complications

Overall complication rates showed a non-significant trend toward higher morbidity in octogenarians, occurring in 30.0% versus 15.0% of younger patients (p = 0.171, absolute risk difference 15.0%, 95% CI: −4.3% to 34.3%) (

Table 4. Postoperative Outcomes (Post-Matching).

Variable	<80 Years (n = 40)	≥80 Years (n = 20)	p-Value	ARD (95% CI)
Any complication, n (%)	6 (15.0)	6 (30.0)	0.171	15.0% (−4.3% to 34.3%)
Clavien–Dindo grade
- Grade I	4 (10.0)	2 (10.0)	1.000	0.0% (−13.9% to 13.9%)
- Grade II	2 (5.0)	2 (10.0)	0.584	5.0% (−8.0% to 18.0%)
- Grade IIIa	0 (0)	1 (5.0)	0.333	5.0% (−4.5% to 14.5%)
- Grade IIIb	0 (0)	1 (5.0)	0.333	5.0% (−4.5% to 14.5%)
- Grade IV	0 (0)	0 (0)	1.000	-
- Grade V	0 (0)	0 (0)	1.000	-
Major complications (≥IIIa), n (%)	0 (0)	2 (10.0)	0.109	10.0% (−3.2% to 23.2%)
Specific complications
- Prolonged air leak (>5 days)	4 (10.0)	3 (15.0)	0.674	5.0% (−10.0% to 20.0%)
- Pneumonia	2 (5.0)	2 (10.0)	0.584	5.0% (−8.0% to 18.0%)
- Atrial fibrillation	2 (5.0)	3 (15.0)	0.315	10.0% (−5.1% to 25.1%)
- Delirium (CAM positive)	0 (0)	4 (20.0)	0.011	20.0% (2.5% to 37.5%)
- Respiratory failure	0 (0)	1 (5.0)	0.333	5.0% (−4.5% to 14.5%)
- Myocardial infarction	0 (0)	0 (0)	1.000	-
- Stroke	0 (0)	0 (0)	1.000	-
- Pulmonary embolism	1 (2.5)	0 (0)	0.999	−2.5% (−7.3% to 2.3%)
Hospital stay, days	6.5 (5–8)	9.5 (7–13)	0.008	-
ICU admission, n (%)	2 (5.0)	3 (15.0)	0.315	10.0% (−5.1% to 25.1%)
ICU stay (if admitted), days	1 (1–1)	2 (1–3)	0.221	-
Chest tube duration, days	3 (2–4)	4 (3–6)	0.042	-
Opioid use, MME	45 (30–60)	30 (20–45)	0.032	-
Discharge disposition			0.024
- Home	38 (95.0)	15 (75.0)
- Rehabilitation facility	2 (5.0)	5 (25.0)
Readmission within 30 days, n (%)	2 (5.0)	3 (15.0)	0.315	10.0% (−5.1% to 25.1%)

ARD: Absolute Risk Difference; MME: Morphine Milligram Equivalents.

). Major complications (Clavien–Dindo grade ≥ IIIa) occurred exclusively in octogenarians (10.0% vs. 0%, p = 0.109), comprising one grade IIIa complication requiring pleural drainage and one grade IIIb complication necessitating surgical re-intervention. The most common complication in both groups was prolonged air leak exceeding 5 days (15.0% vs. 10.0%, p = 0.674). Notably, delirium diagnosed by CAM criteria occurred exclusively in octogenarians (20.0% vs. 0%, p = 0.011, absolute risk difference 20.0%, 95% CI: 2.5% to 37.5%), representing the only statistically significant difference in specific complications (Table 4).

3.6. Hospital Course and Resource Utilization

Octogenarians experienced significantly longer hospital stays with a median of 9.5 days (IQR 7–13) compared to 6.5 days (IQR 5–8) in younger patients (p = 0.008) (Table 4). Chest tube duration was similarly prolonged at 4 days (IQR 3-6) versus 3 days (IQR 2-4) (p = 0.042). Despite higher complication rates, octogenarians required significantly less postoperative opioid analgesia with median consumption of 30 MME (IQR 20-45) compared to 45 MME (IQR 30–60) in younger patients (p = 0.032). ICU admission occurred in 15.0% of octogenarians versus 5.0% of younger patients (p = 0.315), with those admitted requiring longer ICU stays (median 2 vs. 1 day, p = 0.221). Discharge disposition differed significantly between groups (p = 0.024), with 25.0% of octogenarians requiring rehabilitation facility placement compared to only 5.0% of younger patients. The 30-day readmission rate showed a non-significant trend toward higher rates in octogenarians (15.0% vs. 5.0%, p = 0.315) (Table 4).

3.7. Mortality Outcomes

No perioperative mortality occurred in either group, with 30-day mortality of 0% (

Table 5. Mortality Outcomes (Post-Matching).

Outcome	<80 Years (n = 40)	≥80 Years (n = 20)	p-Value	ARD (95% CI)
30-day mortality, n (%)	0 (0)	0 (0)	1.000	-
90-day mortality, n (%)	0 (0)	1 (5.0)	0.333	5.0% (−4.5% to 14.5%)
One-year mortality, n (%)	0 (0)	3 (15.0)	0.035	15.0% (0.6% to 29.4%)

). One octogenarian died within 90 days (5.0% vs. 0%, p = 0.333, absolute risk difference 5.0%, 95% CI: −4.5% to 14.5%), attributed to COVID-19 pneumonia at 75 days postoperatively. One-year mortality was significantly higher in octogenarians at 15.0% (3/20) versus 0% in younger patients (p = 0.035, absolute risk difference 15.0%, 95% CI: 0.6% to 29.4%). The three deaths occurred at 75, 168, and 287 days postoperatively from COVID-19 pneumonia, congestive heart failure, and stroke, respectively. None of these deaths were attributed to surgical complications or cancer recurrence (Table 5).

The three deaths in the octogenarian group occurred at 75, 168, and 287 days postoperatively. Causes were: (1) COVID-19 pneumonia at 75 days, (2) congestive heart failure at 168 days, and (3) stroke at 287 days. None were directly related to surgical complications or cancer recurrence.

4. Discussion

This propensity-matched analysis demonstrates that U-VATS can be performed with acceptable short-term safety in carefully selected octogenarians, despite their reduced pulmonary reserve and higher comorbidity burden. The absence of 30-day mortality and comparable perioperative outcomes suggest that chronological age alone should not preclude consideration for minimally invasive lung resection. A fundamental source of selection bias inherent to this study design warrants explicit acknowledgment. The octogenarian cohort analyzed here represents patients who, following clinical evaluation and multidisciplinary discussion, were deemed physiologically suitable for resection—a prerequisite that excluded the majority of patients aged ≥ 80 years presenting with lung cancer at our institution. Consequently, the reported outcomes apply specifically to a subgroup with sufficient cardiopulmonary reserve and acceptable operative risk and cannot be directly extrapolated to the broader octogenarian lung cancer population. Similarly, all procedures were performed by a single experienced surgeon at a medium-volume center; outcomes achieved under these conditions may not be reproducible across institutions with differing surgical volumes, experience levels, or referral patterns.

The significantly lower FEV1 and DLCO values observed in our octogenarian cohort merit particular attention. Despite matching for major comorbidities, octogenarians demonstrated FEV1 values of 75.2% predicted compared to 87.5% in younger patients, with similar reductions in DLCO. This finding aligns with the well-established physiological aging processes comprehensively described by Janssens and colleagues [12], who documented progressive decline in respiratory muscle strength, chest wall compliance, and alveolar surface area as normal consequences of aging. The preservation of adequate surgical outcomes despite these reduced parameters suggests that traditional pulmonary function thresholds for operability may require reconsideration in elderly populations. Our experience indicates that FEV1 values of 70–80% predicted may represent excellent physiological reserve for octogenarians, particularly when considered within the context of their age-adjusted norms rather than absolute cutoffs designed for younger populations. This interpretation is supported by Oelsner et al. [7], who demonstrated that lung function decline follows predictable age-related trajectories that should inform clinical decision-making.

The technical feasibility of U-VATS in octogenarians is reinforced by our operative data. Comparable operative times, minimal blood loss, and a zero conversion rate demonstrate that advanced age does not inherently increase technical complexity when procedures are performed by experienced surgeons. This contrasts markedly with early VATS literature, which frequently reported prolonged operative times and increased conversion rates in elderly patients [11,13]. However, feasibility must be understood as a multidimensional construct. While procedural execution appears technically achievable in this population, the postoperative recovery trajectory differed substantially: octogenarians experienced significantly higher rates of delirium, longer hospital stay, prolonged chest tube duration, and greater rehabilitation requirements. These findings indicate that although U-VATS can be technically performed in the very elderly, the postoperative course is more resource-intensive and functionally demanding. Future studies should explicitly separate intraoperative technical feasibility from postoperative recovery feasibility as distinct outcome domains when evaluating surgical approaches in elderly populations. The evolution of surgical instruments, refined techniques, and accumulated experience has evidently overcome many of the technical challenges previously associated with minimally invasive surgery in the elderly [14]. Our zero conversion rate compares favorably to contemporary multicenter studies reporting conversion rates of 5–15% in elderly populations [1,2], suggesting that careful patient selection combined with technical expertise can achieve excellent intraoperative outcomes regardless of patient age. The successful implementation of uniportal techniques in this population aligns with the experiences of Abu Akar and colleagues [6], who demonstrated that advanced minimally invasive approaches can be safely applied to complex thoracic procedures.

The postoperative recovery patterns observed in our study reveal important considerations for perioperative management of octogenarians. While the overall complication rate did not reach statistical significance, the trend toward increased morbidity in octogenarians, particularly the occurrence of major complications exclusively in this group, warrants careful interpretation. The absolute risk difference of 10% for major complications, though not statistically significant, may represent clinically meaningful differences that our sample size was insufficient to detect definitively. This finding parallels the meta-analysis by Li et al. [1], which demonstrated that while minimally invasive approaches reduce complications compared to open surgery, elderly patients remain at higher risk for postoperative morbidity regardless of surgical approach [15].

Postoperative delirium emerged as the sole statistically significant complication difference between groups, occurring exclusively in octogenarians (20.0% vs. 0%, p = 0.011). This finding carries substantial clinical and health-economic implications. Delirium in surgical patients is associated with prolonged hospitalization, increased healthcare costs, and accelerated cognitive decline. In our cohort, the association between delirium and extended length of stay (median 9.5 vs. 6.5 days, p = 0.008) and increased discharge to rehabilitation facilities (25.0% vs. 5.0%, p = 0.024) suggests a mechanistic pathway whereby delirium mediates downstream resource utilization.

Leslie et al. [16] estimated that delirium adds approximately $2500 per hospitalization in direct costs, with additional indirect costs from prolonged rehabilitation and potential long-term cognitive sequelae. Extrapolating to our population, the 20% delirium incidence represents a modifiable target for quality improvement initiatives. Multicomponent delirium prevention protocols, including the Hospital Elder Life Program (HELP), have demonstrated 30–40% relative risk reductions in surgical populations [17] and warrant consideration for routine implementation in elderly U-VATS candidates.

The absence of delirium in the younger cohort, while potentially reflecting true age-related vulnerability, may also be influenced by differential surveillance intensity. Nonetheless, our findings reinforce that delirium prevention should constitute a central pillar of perioperative care pathways for octogenarians undergoing minimally invasive thoracic surgery.

The paradoxically lower opioid consumption in octogenarians (30 vs. 45 MME, p = 0.032) likely reflects multiple converging factors. Age-related pharmacokinetic changes, including reduced hepatic clearance and volume of distribution, necessitate lower dosing to achieve equivalent analgesia. Furthermore, elderly patients may underreport pain due to generational attitudes, cognitive changes, or fear of addiction. Alternatively, heightened clinician vigilance regarding opioid-related adverse effects (respiratory depression, delirium) in the elderly may have driven more conservative prescribing practices. Pergolizzi et al. [18] have highlighted that opioid requirements decrease approximately 10–15% per decade after age 40, consistent with our observations. The successful pain management with reduced opioid doses demonstrates that adequate analgesia can be achieved while minimizing the risk of opioid-related adverse effects, particularly cognitive impairment and respiratory depression, which elderly patients are particularly susceptible to. Resource utilization patterns differed significantly between groups, with implications for healthcare planning and patient counseling. The median hospital stay of 9.5 days for octogenarians compared to 6.5 days for younger patients reflects not merely medical complexity but also the intersection of physical recovery, functional capacity, and social support requirements. The finding that 25% of octogenarians required rehabilitation facility placement compared to only 5% of younger patients emphasizes the need for proactive discharge planning beginning in the preoperative period. Early involvement of social services, physical therapy assessment, and family education regarding postoperative care requirements can facilitate smoother transitions and potentially reduce overall length of stay. These findings align with the VIOLET trial [19], which emphasized the importance of functional recovery metrics in evaluating surgical outcomes.

The mortality outcomes in our study require careful contextual interpretation. The absence of 30-day mortality in both groups represents a significant achievement, particularly when compared to historical data reporting perioperative mortality rates of 5–10% for octogenarians undergoing lung resection [3] and likely reflects the cumulative benefits of minimally invasive techniques, enhanced perioperative care protocols, and careful patient selection. The 15% one-year mortality observed in octogenarians, compared with 0% in younger patients (p = 0.035), should not be interpreted as reflecting surgical risk or perioperative harm. All three deaths were attributable to non-surgical, non-oncological causes—COVID-19 pneumonia, congestive heart failure, and stroke—occurring at 75, 168, and 287 days postoperatively. This pattern is consistent with the well-established predominance of competing non-cancer mortality in elderly surgical populations and reflects the background mortality trajectory of octogenarians with high comorbidity burden, a trajectory that persists regardless of surgical intervention. Critically, the higher Charlson Comorbidity Index in the octogenarian group (5.3 vs. 3.8) remained incompletely balanced after matching (SMD = 1.152) and likely contributes substantially to this mortality differential. Conclusions regarding surgical safety must therefore be explicitly decoupled from one-year all-cause mortality in this context. This pattern mirrors findings from Berry et al. [8], who developed predictive models demonstrating that non-cancer mortality becomes increasingly important in elderly surgical patients.

These results are consistent with earlier reports showing one-year mortality rates up to 20% in octogenarians undergoing lung resections [7,20], with most deaths attributed to non-cancer causes [19]. Moreover, our observed delirium rate (20%) aligns with recent findings in thoracic surgery populations using standardized screening protocols [9,10]. The longer median length of stay of 9.5 days is also within the range reported in large trials and retrospective cohorts examining VATS in the elderly [5,19].

These findings have important implications for informed consent discussions. While surgeons can reassure octogenarian candidates that immediate surgical risks are comparable to those in younger patients, the conversation must acknowledge the broader context of competing mortality risks. The discussion should emphasize that successful surgery represents one component of overall health maintenance in elderly patients, and that ongoing medical management of comorbidities remains crucial for long-term survival. This comprehensive approach to patient counseling is consistent with recommendations from recent international consensus statements on cardiothoracic surgery in the elderly [21,22].

The strengths of our study include the use of propensity score matching to reduce, though not eliminate, selection bias, standardized surgical technique by an experienced surgeon, and comprehensive outcome reporting including patient-centered metrics such as discharge disposition and functional recovery. The granular classification of complications using the Clavien–Dindo system provides clinical context that enhances the interpretability of our findings. Furthermore, our setting in a medium-volume center without access to alternative treatments such as SBRT may better represent the reality faced by many thoracic surgical units worldwide, enhancing the generalizability of our findings to similar healthcare environments. For clarity, our center performs approximately 50–60 U-VATS anatomical lung resections per year, placing it within the medium-volume classification by current European Thoracic Society benchmarks.

Nevertheless, several methodological limitations warrant careful consideration. Despite propensity score matching, standardized mean differences remained elevated for the Charlson Comorbidity Index (SMD = 1.152), FEV1 (SMD = 0.726), and DLCO (SMD = 0.760), all exceeding the conventional acceptability threshold of 0.10–0.20. This represents a fundamental limitation of the matching strategy employed: the inclusion of pulmonary function parameters and comorbidity burden directly into the propensity model—or application of inverse probability weighting—may have achieved superior balance. As a consequence, outcome differences between groups, particularly one-year mortality, cannot be attributed exclusively to age-related effects and likely reflect, at least in part, residual confounding by these incompletely balanced variables. All group comparisons should be interpreted within this constraint.

The matched octogenarian cohort comprised only 20 patients, rendering the study substantially underpowered for detecting moderate but clinically meaningful differences in complication rates. Each event in this group corresponds to a 5% rate change, introducing statistical fragility whereby a single additional event would materially alter most estimates. Furthermore, the matching process, by design, selects the most comparable octogenarian candidates, producing a cohort more physiologically homogeneous than the broader operated elderly population. These results should therefore be interpreted as hypothesis-generating, applicable to a highly selected subgroup rather than generalizable to all octogenarian lung resection candidates.

Baseline frailty and functional status were not systematically quantified using validated instruments in all patients, nor were these variables incorporated into the propensity model or outcome adjustment. Comprehensive Geriatric Assessment was performed in approximately 70% of octogenarians; the absence of complete, standardized frailty data precludes adjustment for functional vulnerability as a distinct contributor to postoperative delirium, prolonged hospitalization, and rehabilitation requirements. Whether the observed differences in these outcomes reflect chronological age per se or underlying differences in physiological reserve—independently of the variables captured by the Charlson Comorbidity Index—cannot be determined from the present analysis.

The single-surgeon nature of our experience, while ensuring technical consistency, raises questions about generalizability. Centers with different surgical volumes, experience levels, or patient populations may not achieve comparable outcomes. Additionally, the absence of long-term oncological data, including 5-year survival rates and recurrence patterns, prevents full assessment of whether the oncological adequacy of resection is maintained in elderly patients, particularly given the trend toward slightly less extensive lymph node dissection observed in our octogenarian cohort. This limitation is particularly relevant given recent evidence from Cao et al. [13] suggesting that oncological outcomes may vary with extent of resection in elderly patients. Finally, nearest-neighbor matching without replacement was selected for its interpretability and established performance in small-to-moderate samples; however, stricter caliper widths or alternative approaches such as inverse probability weighting may have achieved superior covariate balance and should be considered in future studies with larger cohorts.

Looking forward, our findings suggest several avenues for future research and clinical development. Prospective studies incorporating comprehensive geriatric assessment tools, including validated frailty indices, cognitive evaluation, and functional capacity measurements, could refine selection criteria for octogenarian surgical candidates. Investigation of enhanced recovery protocols specifically tailored to elderly patients, perhaps incorporating prehabilitation programs, might further improve outcomes. Economic analyses comparing the cost-effectiveness of surgical intervention versus alternative treatments in octogenarians would inform healthcare policy decisions, particularly as the elderly population continues to expand globally.

In this exploratory propensity-matched analysis, U-VATS demonstrated acceptable technical feasibility in selected octogenarians, with comparable intraoperative parameters to younger patients. However, postoperative recovery differed meaningfully, with higher delirium rates, longer hospitalization, prolonged chest tube duration, and greater rehabilitation requirements in the elderly cohort. These findings must be interpreted cautiously given residual imbalances in comorbidity burden (CCI) and pulmonary function (FEV1, DLCO) after matching, which preclude attribution of outcome differences to age alone. One-year mortality, while higher in octogenarians, reflects competing comorbid risk rather than surgical harm and should not be conflated with perioperative safety. Comprehensive geriatric assessment and structured delirium prevention protocols should be integral components of perioperative care pathways for this population. Larger multicenter prospective studies with longer follow-up are warranted to validate these hypothesis-generating findings and establish evidence-based selection criteria for octogenarian U-VATS candidates.