Incidence, risk factors, and clinical impact of subcutaneous emphysema following laparoscopic totally extraperitoneal inguinal herniorrhaphy: a retrospective cohort study

Abstract

Background

We aimed to determine the incidence of subcutaneous emphysema following laparoscopic totally extraperitoneal (TEP) inguinal herniorrhaphy and to evaluate its associated factors and clinical impact.

Methods

In this retrospective cohort study, we included consecutive patients who underwent laparoscopic TEP inguinal herniorrhaphy between 2020 and 2023. The primary outcome was any SE, assessed using palpation method and radiographic imaging. Secondary outcomes were severe SE, defined as the SE extending to the area of the nipple and above. We also evaluated key postoperative recovery indices, comparing patients with and without SE.

Results

Among the 829 included patients, 161 (19.4%) developed any SE, and 98 (11.8%) developed severe SE. Four independent risk factors were identified for any SE: lower body mass index [adjusted odds ratios (aORs): 0.84, 95% confidence interval (CI): 0.77–0.90], higher peak airway pressure (aOR: 4.98, 95% CI: 1.53–16.17, on base-2 logarithmic scale), higher peak partial pressure of end-tidal carbon dioxide (aOR: 250.18, 95% CI: 73.41–852.59, on base-2 logarithmic scale), and use of volume-controlled ventilation (pressure- vs. volume-controlled ventilation: aOR: 0.40, 95% CI: 0.22–0.75) (c-statistic = 0.790). The safe threshold for any SE was peak airway pressure < 25 cm H₂O and peak partial pressure of end-tidal carbon dioxide < 45 mm Hg. Patients with SE showed a higher need for supplemental oxygen after surgery.

Conclusions

Independent factors associated with SE following laparoscopic TEP herniorrhaphy were identified and may facilitate timely recognition and management. SE can be accompanied by transient respiratory changes but is generally self-limited.

Background

Inguinal hernia is a common disease with an estimated prevalence of 7.7% worldwide. [1] More than 20 million patients undergo surgical repair of inguinal hernia annually [2]. Compared to the traditional open repair method, laparoscopic herniorrhaphy offers faster recovery, shorter hospital stays, and reduced pain for patients [3, 4]. Laparoscopic totally extraperitoneal (TEP) approach is one of the leading surgical methods for this procedure [5]. The TEP method carries a lower risk for visceral injury and is recommended by the European Hernia Society as the preferred approach [6]. Serious complications after the procedure are uncommon and include vascular injury, intestinal injury, bowel obstruction, and subcutaneous emphysema (SE) [4, 7, 8]. Although SE is typically a self-limited clinical condition, massive and progressive SE following laparoscopy can be life-threatening, potentially leading to expanding pneumothorax, tracheal compression, and respiratory and cardiovascular compromise [9–11].

The incidence of SE following laparoscopic surgery varies significantly across studies, ranging from 2.3-99%.^12–14 This variation in reported incidence may be due to differences in study populations, diagnostic criteria, or surgical methods. A retrospective cohort study examined potential risk factors for hypercarbia-related complications in laparoscopy, revealing that a partial pressure of end-tidal carbon dioxide (PETCO₂) greater than 50 mm Hg, prolonged operative duration, and the use of six or more surgical ports may be associated with increased SE.¹⁴ This study included various procedures instead of focusing solely on laparoscopic herniorrhaphy [14]. In another prospective cohort study on adults undergoing laparoscopic inguinal hernia surgery, a higher body mass index (BMI), longer surgical duration, and elevated PETCO₂ were associated with an increased incidence of SE.¹³ However, the generalizability of this study was limited by its relatively small sample size, the lack of multivariable analysis, and the inclusion of only one surgeon [13].

In this single-center retrospective study, we aimed to determine the incidence of SE following laparoscopic TEP hernia repair and to explore potential risk factors for SE. This study also assessed the impact of SE on postoperative recovery, a largely unexplored area in previous research [12–14]. The results may provide insights into modifiable factors that could aid in the early identification of high-risk patients and prevent the occurrence of SE and its associated complications.

Methods

Inclusion and exclusion criteria

We obtained the approval from the Taipei Medical University Joint Institutional Review Board (TMU-JIRB number: N202411041). The need for informed consent was waived by the Institutional Review Board due to the retrospective nature of this research. This study was performed in accordance with the ethical standards of the Helsinki Declaration 2013 and the STROBE Reporting Guidelines [15].

Patients who underwent laparoscopic TEP inguinal hernia repair at Shuang Ho Hospital, Taipei Medical University from January 1, 2020, to December 31, 2023 were consecutively included for analysis. Exclusion criteria were reoperation cases, critical missing data, age under 18 years, presence with an obstructed or strangulated hernia, combined surgery with non-hernia procedures, undergoing mechanical ventilation, laparoscopic or robotic surgery within 30 days prior to the surgery, and the use of supraglottic airways for general anesthesia.

Surgical and anesthetic management

Al laparoscopic TEP herniorrhaphies were performed by experienced general surgeons with over 200 cases each. Access to the extraperitoneal space was achieved through a 2-cm infraumbilical incision placed just lateral to the midline on the side of the hernia. This space was developed using a 10-mm zero-degree telescope, following the retraction of the rectus muscle from the medial to the lateral aspect. Throughout the procedure, carbon dioxide pressure was typically maintained at 10–15 mm Hg. TEP repair was completed using three midline ports. The TEP procedure was performed in a standard manner utilizing three ports, as outlined by Jorge Daes [16]. A 0° telescope was employed for blunt dissection to establish the pre-peritoneal space. The hernial sac was meticulously dissected and ligated using a catgut 2 − 0 endo-loop. Adequate space was prepared lateral to the internal inguinal ring for the insertion of a mesh, ensuring it lay flat without folds or wrinkles. Mesh fixation was employed only in the presence of a large defect or a sliding hernia. The contralateral side was routinely inspected for hernia presence, and mesh placement was performed without additional ports if necessary.

General anesthesia was performed according to routine institutional practice. Induction typically involved fentanyl (1–2 µg·kg⁻¹) and propofol (1–2 mg·kg⁻¹), with neuromuscular blockade achieved using rocuronium (0.6–1.0 mg·kg⁻¹) or cisatracurium (0.1–0.2 mg·kg⁻¹) to facilitate endotracheal intubation. Mechanical ventilation was delivered via an endotracheal tube (internal diameter 6.5–8.0 mm), using pressure-controlled ventilation (PCV; peak pressure < 30 cm H₂O) or volume-controlled ventilation (VCV; tidal volume 6–8 mL·kg⁻¹ predicted body weight with positive end-expiratory pressure of 4–6 cm H₂O), at the discretion of the attending anesthesiologist, with the goal of maintaining end-tidal carbon dioxide below 45 mm Hg. Volume-guaranteed PCV (PCV-VG; tidal volume 6 mL·kg⁻¹ predicted body weight, peak pressure < 30 cm H₂O) was also used when clinically indicated to optimize ventilation and oxygenation. The fraction of inspired oxygen was adjusted (0.5–0.7) to maintain peripheral oxygen saturation above 95%. Anesthesia was maintained with sevoflurane (0.6–2.5%) without nitrous oxide. At the end of the surgery, neuromuscular blockade was reversed with sugammadex (2 mg·kg⁻¹) or neostigmine (0.03–0.07 mg·kg⁻¹) plus glycopyrrolate, consistent with standard care.

Study outcomes

The primary outcome was any SE following laparoscopic TEP herniorrhaphy. As part of routine postoperative care, patients were clinically evaluated for SE by certified nurse anesthetists immediately after endotracheal extubation and again at approximately 8 h postoperatively. Assessment was based on standard physical examination findings, including crepitus, distention, or bloating on palpation to determine the presence and extent of SE. Radiographic imaging was obtained selectively at the discretion of the clinical team when SE was suspected, with imaging findings defined by areas of radiolucency or a characteristic fluffy appearance along the chest or abdominal wall borders [17]. The secondary outcome was severe SE after surgery, defined as the SE extending to the area of the nipple and above [13]. We also investigated the potential impact of SE on recovery after laparoscopic hernioplasty. Outcomes included post-anesthesia care unit (PACU) length of stay, nadir peripheral oxygen saturation, supplemental oxygen, and accumulative morphine consumption at the PACU. Additionally, we assessed the incidence of postoperative nausea and vomiting, peak daily numeric rating scale (NRS) pain scores within 60 postoperative hours, need for opioid analgesia in the ward, and postoperative length of hospital stay.

Covariates for adjustment

We used the electronic medical record system to collect data and adjust for potential confounders based on data availability, physiological plausibility, and existing literature [12–14, 18, 19]. The demographic data were age, sex, and BMI. Preoperative clinical features included the American Society of Anesthesiologists physical status classification, current smoking (within 30 days before surgery), coexisting diseases (hypertension, diabetes mellitus, asthma, and chronic obstruction pulmonary disease), and preoperative laboratory testing (hemoglobin, creatinine, and estimated glomerular filtration rate based on the Cockcroft-Gault formula) [20]. The surgical and anesthetic factors included hernia side (unilateral or bilateral), hernia classification (indirect, direct, or mixed), duration and pressure of abdominal insufflation, number of ports, endotracheal tube size, ventilator mode during surgery (VCV, PCV, and/or PCV-VG), peak airway pressure during abdominal insufflation, baseline and peak PETCO₂ levels during preperitoneal insufflation, duration of anesthesia, intraoperative estimated blood loss, and intravenous fluid volume.

Statistical analysis

Shapiro–Wilk and Kolmogorov–Smirnov tests were used to check the normality of the collected variables. A logarithmic transformation was applied to decrease the distribution skewness of non-normally distributed factors in the univariate and multivariable models, including preoperative laboratory testing, peak airway pressure, abdominal insufflation duration and pressure, and baseline and peak levels of PETCO₂. Chi-squared or Fisher’s exact tests were used to compare the distributions of baseline patient characteristics and postoperative recovery indices between patients with and without SE for categorical variables, and either independent t- or Wilcoxon rank-sum tests for continuous data, as appropriate. We used logistic regression analyses to evaluate the association between the collected variables and SE. Significant factors identified by univariate analyses were incorporated into the multivariable backward stepwise regression models with the entry criteria of a significance level of 0.05 and exit criteria of 0.10 to determine independent explanatory factors and to calculate adjusted odds ratios (aORs) and 95% confidence intervals (CIs) for any and severe SE. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUROC) for all patients, patients aged ≥ 65 or < 65 years, and males or females. A two-sided significance level of 0.05 was used to define a statistically significant difference. Statistical analyses were performed using the SAS software (version 9.4; SAS Institute Inc., Cary, NC, USA).

Results

Baseline patient characteristics

After the selection process, a total of 829 patients were included in the analysis (Fig. 1). The baseline characteristics of the patients are presented in Table 1. Notably, patients with SE had a lower BMI, higher peak airway pressure and peak levels of PETCO₂ during preperitoneal insufflation, as well as longer duration and higher pressure of preperitoneal insufflation compared to those without SE. Concerning anesthesia parameters, patients who developed SE were less likely to use VCV alone, more likely to use VCV combined with PCV during surgery, and experienced a longer duration of anesthesia.

Fig. 1.

Flow diagram for patient selection. TEP = totally extraperitoneal

Table 1.

Baseline characteristics of the included patients

	Subcutaneous emphysema (n = 161)	No subcutaneous emphysema (n = 668)	p	All patients (n = 829)
Age, year	64.1 ± 12.1	62.2 ± 14.8	0.0866	62.6 ± 14.3
Sex, male	152 (94.4%)	643 (96.4%)	0.2462	795 (96.0%)
Body mass index, kg·m− 2	23.2 ± 2.9	24.3 ± 3.2	< 0.0001	24.1 ± 3.2
ASA class			0.3005
I	30 (18.6%)	159 (23.8%)		189 (22.8%)
II	130 (80.8%)	505 (75.6%)		635 (76.6%)
III	1 (0.6%)	4 (0.6%)		5 (0.6%)
Current cigarette smoking	42 (26.1%)	166 (24.9%)	0.7453	208 (25.1%)
Hypertension	45 (28.0%)	226 (33.8%)	0.1532	271 (32.7%)
Diabetes mellitus	16 (9.9%)	80 (12.0%)	0.4682	96 (11.6%)
Asthma	7 (4.4%)	16 (2.4%)	0.1757	23 (2.8%)
COPD	4 (2.5%)	10 (1.5%)	0.4914	14 (1.7%)
Preoperative laboratory testing
Hemoglobin, g·dL− 1	14.6 (13.7–15.2)	14.5 (13.6–15.4)	0.6947	14.5 (13.6–15.3)
Creatinine, mg·dL− 1	0.93 (0.83–1.10)	0.92 (0.82–1.06)	0.2387	0.92 (0.82–1.06)
eGFR, mL·min·1.73 m− 2	87.6 (72.5–99.7)	89.5 (75.7–101.5)	0.1284	89.1 (74.9–100.9)
Side of hernia			0.1723
Unilateral	93 (57.8%)	424 (63.6%)		517 (62.4%)
Bilateral	68 (42.2%)	243 (36.4%)		311 (37.6%)
Classification of hernia			0.2101
Indirect	95 (59.4%)	398 (59.9%)		493 (59.8%)
Direct	28 (17.5%)	141 (21.2%)		169 (20.5%)
Mixed	11 (6.9%)	54 (8.1%)		65 (7.9%)
Other†	26 (16.3%)	71 (10.7%)		97 (11.8%)
Endotracheal tube size, mm			0.8693
6.5	6 (0.9%)	1 (0.6%)		7 (0.8%)
7.0	57 (8.5%)	15 (9.3%)		72 (8.7%)
7.5	594 (88.9%)	144 (89.4%)		738 (89.0%)
8.0	11 (1.7%)	1 (0.6%)		12 (1.5%)
Peak airway pressure, cm H2O	26 (25–29)	26 (23–28)	0.0021	26 (23–28)
Port number			0.0894
One	3 (1.9%)	7 (1.1%)		10 (1.2%)
Two	1 (0.6%)	1 (0.2%)		2 (0.3%)
Three	154 (96.9%)	641 (98.8%)		795 (98.4%)
Four	1 (0.6%)	0 (0.0%)		1 (0.1%)
Abdominal insufflation duration, min	55 (42–75)	51 (37–70)	0.0218	52 (38–71)
Abdominal insufflation pressure, mm Hg	15 (15–15)	15 (15–15)	0.0089	15 (15–15)
PETCO2, mm Hg
Baseline	32 (31–34)	32 (31–34)	0.0799	32 (31–34)
Peak	47 (43–51)	41 (38–44)	< 0.0001	42 (38–46)
Ventilator mode			< 0.0001
VCV	41 (25.5%)	290 (43.5%)		331 (40.0%)
PCV	3 (1.9%)	22 (3.3%)		25 (3.0%)
VCV and PCV	90 (55.9%)	237 (35.6%)		327 (39.5%)
VCV and PCV-VG	27 (16.8%)	117 (17.6%)		144 (17.4%)
Anesthesia duration, min	130 (115–150)	125 (105–145)	0.0232	125 (105–150)
Intraoperative blood loss, mL	10 (10–10)	10 (10–10)	0.1185	10 (10–10)
Intraoperative fluid volume, mL	600 (500–700)	600 (500–700)	0.8967	600 (500–700)

Values are mean ± standard deviation, median (interquartile range) or counts (percent). 

ASA American Society of Anesthesiologists, COPD chronic obstruction pulmonary disease, eGFR estimated glomerular filtration rate, PCV pressure-controlled ventilation, PCV-VG volume-guaranteed pressure-controlled ventilation, PETCO₂ partial pressure of end-tidal carbon dioxide, VCV volume-controlled ventilation.

† Bilateral inguinal hernias of two different types (e.g., direct type on the right and indirect on the left)

Rate and severity of subcutaneous emphysema

Table 2 illustrates the rate and extent of SE following laparoscopic TEP inguinal hernia repair. Among the included patients, 161 (19.4%) developed any SE, and 98 (11.8%) experienced severe SE. Specifically, severe SE extended to the face, neck, and clavicle in 13 (1.6%), 53 (6.4%), and 32 (3.9%) patients, respectively.

Table 2.

Rates and extent of subcutaneous emphysema after laparoscopic inguinal herniorrhaphy

	All patients (n = 829)
Any subcutaneous emphysema	161 (19.4%)
Severe subcutaneous emphysema	98 (11.8%)
Extent of subcutaneous emphysema
Abdomen, below umbilicus	14 (1.7%)
Abdomen, umbilicus to xiphisternum	8 (1.0%)
Chest wall, xiphisternum to nipple	41 (4.9%)
Chest wall, nipple to clavicle	32 (3.9%)
Neck	53 (6.4%)
Face	13 (1.6%)

Values are counts (percent)

Risk factors of subcutaneous emphysema

Univariate analyses identified several factors significantly associated with any SE, including lower BMI, higher peak airway pressure, longer and higher-pressure abdominal insufflation, elevated peak PETCO₂ levels, and ventilator mode (Table 3). After multivariable adjustment, four independent risk factors remained: lower BMI, higher peak airway pressure, higher peak PETCO₂ levels, and ventilator mode. The associations between peak airway pressure, peak PETCO₂, and the incidence of both any and severe SE are illustrated in Fig. 2, which demonstrates a marked increase in SE rates above a peak airway pressure of 25 cm H₂O and a peak PETCO₂ level of 45 mm Hg. The multivariable model for predicting any SE showed good discrimination (AUROC 0.790; Fig. 3A), with higher discriminatory performance observed in female patients compared with male patients.

Table 3.

Univariate and multivariable analyses for subcutaneous emphysema after laparoscopic hernia repair

	Univariate cOR (95% CI)	p	Multivariable aOR (95% CI)	p
Age, year	1.010 (0.997–1.022)	0.1286	-	-
Sex, male vs. female	0.63 (0.29–1.38)	0.2492	-	-
Body mass index, kg·m− 2	0.89 (0.84–0.94)	< 0.0001	0.84 (0.77–0.90)	< 0.0001
ASA class		0.3758	-	-
II vs. I	1.36 (0.88–2.11)	0.7681	-	-
III vs. I	1.33 (0.14–12.27)	0.9107	-	-
Current cigarette smoking	1.07 (0.72–1.58)	0.7453	-	-
Hypertension	0.76 (0.52–1.11)	0.1541	-	-
Diabetes mellitus	0.81 (0.46–1.43)	0.4689	-	-
Asthma	1.85 (0.75–4.58)	0.1821	-	-
COPD	1.68 (0.52–5.42)	0.3878	-	-
Preoperative laboratory testing†			-	-
Hemoglobin, g·dL− 1	0.72 (0.25–2.09)	0.5410	-	-
Creatinine, mg·dL− 1	1.24 (0.78–1.97)	0.3659	-	-
eGFR, mL·min·1.73 m− 2	0.75 (0.51–1.10)	0.1454	-	-
Side of hernia			-	-
Bilateral vs. unilateral	1.276 (0.90–1.81)	0.1728	-	-
Classification of hernia		0.2157	-	-
Direct vs. indirect	0.83 (0.52–1.32)	0.2725	-	-
Mixed vs. indirect	0.85 (0.43–1.69)	0.4913	-	-
Other vs. indirect	1.53 (0.93–2.53)	0.0418	-	-
Endotracheal tube size, mm		0.7816	-	-
7.0 vs. 6.5	1.58 (0.18–14.14)	0.3571	-	-
7.5 vs. 6.5	1.46 (0.17–12.18)	0.4130	-	-
8.0 vs. 6.5	0.55 (0.03–10.37)	0.4260	-	-
Peak airway pressure, cm H2O†	3.70 (1.50–9.16)	0.0046	4.98 (1.53–16.17)	0.0076
Port number		0.6388	-	-
Two vs. one	2.33 (0.11–50.98)	0.9856	-	-
Three vs. one	0.56 (0.14–2.19)	0.9777	-	-
Four vs. one	> 999.99 (< 0.01–999.99)	0.9814	-	-
Abdominal insufflation duration, min†	1.35 (1.02–1.78)	0.0364	-	-
Abdominal insufflation pressure, mm Hg†	7.98 (1.69–37.57)	0.0086	-	-
PETCO2, mm Hg †			-	-
Baseline	0.11 (0.01–1.02)	0.0520	-	-
Peak	218.01 (78.08–608.67)	< 0.0001	250.18 (73.41–852.59)	< 0.0001
Ventilator mode		< 0.0001	-	0.0247
PCV vs. VCV	0.97 (0.28–3.37)	0.3963	0.40 (0.22–0.75)	0.0039
VCV and PCV vs. VCV	2.69 (1.79–4.04)	0.0011	0.33 (0.08–1.37)	0.1272
VCV and PCV-VG vs. VCV	1.63 (0.96–2.78)	0.5702	0.63 (0.35–1.14)	0.1297

aOR adjusted odds ratio, ASA American Society of Anesthesiologists, CI confidence interval, cOR crude odds ratio, COPD chronic obstruction pulmonary disease, eGFR estimated glomerular filtration rate, PCV pressure-controlled ventilation, PCV-VG volume-guaranteed pressure-controlled ventilation, PETCO₂ partial pressure of end-tidal carbon dioxide, VCV volume-controlled ventilation. 

† On base-2 logarithmic scale

Fig. 2.

Rates of any and severe subcutaneous emphysema stratified by peak airway pressure (A; unit, cm H₂O) and peak partial pressure of end-tidal carbon dioxide during preperitoneal insufflation (B; unit, mm Hg). SE = subcutaneous emphysema

Fig. 3.

Receiver operating characteristic curves of the multivariable models for any subcutaneous emphysema (A1 to A5) and for severe subcutaneous emphysema (B1 to B5). Values are area under the curve (AUC) and 95% confidence limit

Similarly, four independent predictors of severe SE were identified: lower BMI, higher peak airway pressure, higher peak PETCO₂ levels, and ventilator mode (Table 4). The model for severe SE demonstrated strong overall performance (AUROC 0.825; Fig. 3B), again with higher AUROC values in females than in males.

Table 4.

Univariate and multivariable analyses for severe subcutaneous emphysema after laparoscopic hernia repair

	Univariate cOR (95% CI)	p	Multivariable aOR (95% CI)	p
Age, year	1.013 (0.998–1.029)	0.0998	-	-
Sex, male vs. female	0.40 (0.18–0.91)	0.0292	-	-
Body mass index, kg·m− 2	0.89 (0.83–0.96)	0.0012	0.84 (0.76–0.92)	0.0001
ASA class		0.6734	-	-
II vs. I	1.27 (0.75–2.16)	0.9722	-	-
III vs. I	< 0.01 (< 0.01 – >999.99)	0.9731	-	-
Current cigarette smoking	1.03 (0.63–1.66)	0.9183	-	-
Hypertension	0.72 (0.45–1.15)	0.1679	-	-
Diabetes mellitus	1.08 (0.56–2.05)	0.8267	-	-
Asthma	1.60 (0.53–4.79)	0.4055	-	-
COPD	1.25 (0.28–5.66)	0.7736	-	-
Preoperative laboratory testing†			-	-
Hemoglobin, g·dL− 1	0.68 (0.19–2.49)	0.5609	-	-
Creatinine, mg·dL− 1	1.01 (0.55–1.83)	0.9801	-	-
eGFR, mL·min·1.73 m− 2	0.81 (0.51–1.29)	0.3733	-	-
Side of hernia			-	-
Bilateral vs. unilateral	1.48 (0.97–2.27)	0.0700	-	-
Classification of hernia		0.1880	-	-
Direct vs. indirect	0.93 (0.52–1.65)	0.1788	-	-
Mixed vs. indirect	1.51 (0.73–3.14)	0.5022	-	-
Other vs. indirect	1.76 (0.97–3.20)	0.1378	-	-
Endotracheal tube size, mm		0.6455	-	-
7.0 vs. 6.5	1.20 (0.13–10.89)	0.9804	-	-
7.5 vs. 6.5	0.78 (0.09–6.57)	0.9828	-	-
8.0 vs. 6.5	< 0.01 (< 0.01 – >999.99)	0.9816	-	-
Peak airway pressure, cm H2O†	5.60 (1.79–17.47)	0.0030	6.31 (1.48–26.92)	0.0128
Port number		0.5637	-	-
Two vs. one	9.00 (0.28–285.51)	0.9908	-	-
Three vs. one	1.21 (0.15–9.63)	0.9826	-	-
Four vs. one	> 999.99 (< 0.01 – >999.99)	0.9851	-	-
Abdominal insufflation duration, min†	1.50 (1.06–2.13)	0.0208	-	-
Abdominal insufflation pressure, mm Hg†	5.38 (0.85–34.16)	0.0745	-	-
PETCO2, mm Hg†			-	-
Baseline	0.30 (0.02–4.28)	0.3717	-	-
Peak	316.69 (96.86–1035.46)	< 0.0001	286.83 (70.26–1170.94)	< 0.0001
Ventilator mode		< 0.0001		0.0626
PCV vs. VCV	2.12 (0.59–7.69)	0.8273	0.34 (0.15–0.77)	0.0090
VCV and PCV vs. VCV	3.49 (2.05–5.95)	0.0037	0.72 (0.17–3.10)	0.6611
VCV and PCV-VG vs. VCV	1.81 (0.90–3.64)	0.8311	0.64 (0.31–1.33)	0.2291

aOR adjusted odds ratio, ASA American Society of Anesthesiologists, CI confidence interval, cOR crude odds ratio, COPD chronic obstruction pulmonary disease, eGFR estimated glomerular filtration rate, PCV pressure-controlled ventilation, PCV-VG volume-guaranteed pressure-controlled ventilation, PETCO₂ partial pressure of end-tidal carbon dioxide, VCV volume-controlled ventilation.

† On base-2 logarithmic scale

Clinical impact of subcutaneous emphysema

Table 5 presents postoperative outcomes comparing patients with and without SE. Patients with SE were significantly more likely to require supplemental oxygen at the PACU. Interestingly, those with SE reported significantly lower average peak daily NRS pain scores (1.2 vs. 1.5, p = 0.0089) within the first 60 postoperative hours. However, there were no statistically significant differences between the groups in PACU length of stay, postoperative length of hospital stay, or incidence of postoperative nausea and vomiting.

Table 5.

Postoperative recovery indices in patients with and without subcutaneous emphysema

	Subcutaneous emphysema (n = 161)	No subcutaneous emphysema (n = 668)	p
PACU length of stay, min	47 ± 6	47 ± 7	0.6495
SpO2 nadir, %	97 ± 2	97 ± 2	0.9481
Supplemental oxygen therapy at PACU			0.0035
None	0 (0)	36 (5.4%)
1–5 L·min− 1	153 (95.0%)	615 (92.1%)
6–10 L·min− 1	8 (5.0%)	17 (2.5%)
Morphine consumption at PACU, mg	3.0 (0.0–5.5)	4.0 (0.0–6.0)	0.4398
Postoperative nausea and vomiting	10 (6.2%)	40 (6.0%)	0.9150
Daily maximum NRS pain score
POH 0–12	2.0 (1.0–2.0)	2.0 (1.0–3.0)	0.0112
POH 12–36	1.0 (0.0–2.0)	1.0 (0.0–2.0)	0.0920
POH 36–60	0.5 (0.0–1.0)	1.0 (1.0–2.0)	0.0257
Mean daily maximum NRS pain score	1.2 (0.6–2.0)	1.5 (1.0–2.0)	0.0089
Need for opioid analgesia in the ward	22 (13.7%)	71 (10.6%)	0.2732
Length of hospital stay, day	2 (2–2)	2 (2–2)	0.7010

Values are mean ± standard deviation, median (interquartile range) or counts (percent). 

NRS numerical rating scale, PACU post-anesthesia care unit, POH postoperative hour, SpO₂ peripheral oxygen saturation by pulse oximetry

Discussion

In this retrospective cohort study, the incidence of SE following laparoscopic TEP hernia surgery was 19.4%. Multivariable analysis identified four risk factors associated with the occurrence of SE, including a lower BMI, higher peak airway pressure, higher PETCO₂ peak levels, and the use of VCV mode compared to PCV mode. The identified demographic and intraoperative factors could be valuable for the early identification and prevention of extensive SE and its related complications. We observed an association between SE and increased use of supplemental oxygen therapy, which may reflect transient respiratory changes or proactive clinical management. To our knowledge, this study provides one of the more detailed evaluations of the relationship between SE and postoperative recovery parameters in this setting.

Several previous studies have examined the relationship between laparoscopic surgery and SE.^12–14 However, only a few have specifically focused on inguinal herniorrhaphy [13]. Murdock and colleagues reviewed 968 laparoscopic procedures and found that only 2.3% of patients developed SE.¹⁴ It should be noted that this retrospective study included various types of surgeries, with only 6.9% of the cases being hernia repairs [14]. In another prospective randomized study of gynecologic laparoscopies, SE was detected in 13.5% of patients by postoperative manual palpation [12]. In contrast, Saggar and colleagues reported a staggering 99% incidence of SE in patients following TEP inguinal hernia surgery [13]. The wide variation in reported incidence rates may result from differences in the types of surgeries, patient characteristics, study populations, and diagnostic methods for SE. For example, computed tomography (CT) is known as the most sensitive imaging modality for diagnosing SE. [21] A study diagnosing SE using routine CT may reveal a significantly higher incidence compared to those using manual palpation alone.

Previous researchers have proposed a variety of clinical factors associated with the occurrence of SE, including high intraoperative PETCO₂, longer surgical duration, older patient age, and factors related to surgical equipment [12–14, 22–24]. In our study, the patient’s intraoperative peak PETCO₂ was found to be associated with increased odds of SE occurrence. The elevation of PETCO₂ may be attributed to increased CO₂ absorption through subcutaneous tissue, leading to hypercapnia [14]. Several reports have indicated that a sudden elevation of PETCO₂ may be an early indicator of SE. [8, 25] In our univariable analysis, preperitoneal insufflation pressure, insufflation duration, and anesthesia duration were significantly associated with SE. However, after adjusting for other variables in the multivariable analysis, these associations were no longer significant. There may be potential confounding among the influential factors. For example, operative time might be longer in obese patients [26]. Interestingly, BMI appears to have conflicting effects on the occurrence of SE across different studies [12, 13, 22]. Our study found that a higher BMI was associated with less SE. Ng et al. reported similar findings and suggested that subcutaneous fat might have a protective effect against CO₂-related sequelae after laparoscopy. Alternatively, this association could be due to underdiagnosis, as palpating for crepitus might be more challenging with thicker subcutaneous fat [22]. Conversely, Saggar and colleagues found that a higher BMI may correlate with the occurrence of SE.¹³ However, the specific diagnostic methods used and whether radiographic studies were incorporated for diagnosis were not detailed in their work [13]. Further studies are needed to clarify the potential causal relationship and pathophysiology between BMI and SE.

In our analyses, VCV was associated with the development of SE. During laparoscopic surgeries, increased minute ventilation is commonly employed to counteract the rising hypercapnia from absorbed CO₂ [27]. SE can result in prolonged hypercapnia due to the accumulation of CO₂ in the subcutaneous layer. Compared to VCV, PCV may provide higher tidal volumes, increased minute ventilation, and better CO2 wash-out, while maintaining similar peak airway pressures. However, the protective effect of PCV appeared to diminish when combined with VCV during a single surgery. At our hospital, the anesthetists often switched the ventilator mode from VCV to PCV intraoperatively to address rising hypercapnia after SE was recognized. The protective benefits of PCV might be preventative and could be overshadowed by the increasing CO₂ burden once clinically significant SE occurs. Furthermore, extraperitoneal insufflation and the Trendelenburg position used during TEP surgery present additional challenges to the patient’s respiratory function. In a study by Sharma and colleagues, extraperitoneal insufflation led to decreased pulmonary compliance and increased peak inspiratory pressure in adults undergoing TEP inguinal herniorrhaphy [28]. Therefore, the anesthesia team may use high peak airway pressure to counteract pulmonary changes and maintain adequate minute ventilation in cases of extensive SE.

We also stratified the incidence rates of SE by varying levels of peak airway pressure and peak PETCO₂. In our study, the incidence of SE increased notably in groups with peak airway pressure above 25 cm H2O. The low incidence observed in the group with peak airway pressure above 35 cm H₂O may be attributed to the small number of patients in this group. A significantly increased incidence of both SE of any degree and severe SE was found in groups with PETCO₂ above 45 mm Hg. Hypercapnia is defined as an arterial partial pressure of CO₂ (PaCO₂) exceeding 45 mm Hg. [29] Since the PaCO2 - ETCO₂ gap can range from 0 to 10 mm Hg in anesthetized patients, hypercapnia may present with ETCO₂ levels ranging from 45 to over 55 mm Hg. [30] Saggar et al. reported an increased incidence of SE in patients with peak PETCO₂ exceeding 40 mm Hg.¹³ In the retrospective cohort study by Murdock and colleagues, a PETCO₂ level exceeding 50 mm Hg was identified as an independent risk factor for SE, pneumothorax, or pneumomediastinum [14]. Our results align with previous literature, suggesting that an ETCO₂ level above 45 mm Hg could indicate the development of SE. It should be noted that elevated airway pressure and PETCO₂ should be interpreted as associated clinical markers rather than causal determinants of SE. They may reflect increased CO₂ absorption from extraperitoneal insufflation, impaired ventilation from evolving SE, or compensatory ventilatory adjustments to hypercapnia or reduced pulmonary compliance. Although elevated airway pressure could theoretically promote air dissection via barotrauma-related mechanisms (e.g., the Macklin effect), CO₂ insufflation inherent to the TEP approach remains a plausible primary driver of subcutaneous gas tracking.

Notably, the discriminatory power of our multivariable models for SE is better for women compared to men. Similar sex-based differences were observed in the prediction of severe SE. In a retrospective analysis, Hayashi et al. found that male sex was a preventive factor for clinically relevant SE after robotic gastrectomy [31]. Our results did not demonstrate similar results, which may be attributable to the relatively small number of female participants in our cohort. Although these findings could suggest sex-specific variation, they should be interpreted with caution, as a limited female sample may yield a more homogeneous risk profile, inflate model performance estimates, and raise concerns about overfitting. Accordingly, the observed differences may reflect sample characteristics rather than true biological effects, underscoring the need for larger, sex-balanced studies.

Few studies have examined the impact of SE on postoperative recovery after laparoscopic herniorrhaphy. In our cohort, patients with SE more frequently received supplemental oxygen, which may reflect transient hypercapnia, reduced chest wall compliance, or proactive clinical vigilance leading to empiric oxygen administration rather than overt respiratory failure. Although severe complications such as chest wall restriction or tracheal compression have been described in isolated case reports, [32, 33] SE is typically a self-limited condition, and its clinical consequences in this setting appear modest. Contrary to expectation, patients with SE reported lower postoperative pain scores in the first 60 postoperative hours, although this difference was not clinically significant. This unexpected finding may be attributed to proactive analgesic prescriptions by clinicians for patients who develop SE. Additionally, the similar lengths of PACU and hospital stays and comparable rates of nausea and vomiting between groups indicate that while SE significantly impacted immediate respiratory function, its overall effect on postoperative recovery might be less pronounced.

Based on our findings and previous studies, it is advisable for clinicians to take preventive measures against SE in laparoscopic inguinal herniorrhaphy and act promptly at the earliest signs of hypercapnia. Since VCV is associated with higher odds of developing SE, the anesthesia team should consider prioritizing PCV for better CO₂ elimination, particularly in patients with a lower BMI. Additionally, close monitoring of peak airway pressure is essential throughout the procedure. If both elevated PETCO₂ and high peak airway pressure are present in obese patients, clinicians should consider utilizing other imaging modalities for a more accurate diagnosis.

Our study has several limitations. First, being retrospective, it cannot clarify the causal relationship between various clinical factors and the occurrence of SE. Second, underdiagnosis of SE may have occurred because case ascertainment was based on a composite of clinical assessment and selective radiographic imaging, and not all patients underwent imaging. Third, we did not collect certain surgery-related factors due to incomplete data, such as types of trocars, detailed surgical techniques and approaches, or the surgeons’ level of experience. Fourth, our analysis demonstrated a strong association between elevated peak PETCO₂ and SE. Importantly, because PETCO₂ was modeled on a base-2 logarithmic scale, this estimate reflects the risk associated with a doubling of PETCO₂ levels. The wide confidence interval reflects limited precision, likely due to event rarity and the influence of extreme hypercapnic values. Lastly, as the study was conducted in a single-center setting, it may lack generalizability. Given the high prevalence of inguinal hernia and the significant incidence of SE revealed by this study, large-scale prospective trials would be beneficial.

Conclusions

SE is a common complication following laparoscopic TEP hernia surgery in adults, particularly among patients with a lower BMI, and is associated with increased use of supplemental oxygen. Although SE is typically benign and self-limited, rare cases of clinically significant progression have been reported. Accordingly, clinicians should remain attentive during and after the procedure, monitoring for early indicators of SE, such as increasing airway pressure or rising PETCO₂, and respond promptly to mitigate potential respiratory compromise. Randomized controlled trials are needed to evaluate the benefits of actively managing airway pressure and PETCO₂ to prevent SE after laparoscopic surgery.

Abbreviations

aOR

adjusted odds ratio

AUROC

area under the receiver operating characteristic curve

BMI

body mass index

CI

confidence interval

cOR

crude odds ratio

COPD

chronic obstructive pulmonary disease

eGFR

estimated glomerular filtration rate

NRS

numerical rating scale

PACU

post-anesthesia care unit

PCV

pressure-controlled ventilation

PCV-VG

volume-guaranteed pressure-controlled ventilation

PETCO₂

partial pressure of end-tidal carbon dioxide

POH

postoperative hour

SE

subcutaneous emphysema

SpO₂

peripheral oxygen saturation by pulse oximetry

TEP

totally extraperitoneal

VCV

volume-controlled ventilation

Authors’ contributions

Conceptualization, Jhih-Wei Wang and Ying-Hsuan Tai; methodology, Ying-Hsuan Tai; validation, Hsiang-Ling Wu; formal analysis, Ying-Hsuan Tai; investigation, Yu-Hsiang Chung, Jhih-Wei Wang, and Yu-Chuan Yang; resources, Jui-Tai Chen and Ying-Hsuan Tai; data curation, Yu-Hsiang Chung, Jhih-Wei Wang, and Yu-Chuan Yang; supervision, Juan P. Cata, Yih-Giun Cherng and Ying-Hsuan Tai; project administration, Ying-Hsuan Tai; writing—original draft preparation, Yu-Hsiang Chung; writing—review and editing, Jhih-Wei Wang, Yu-Chuan Yang, Hsiang-Ling Wu, Juan P. Cata, Jui-Tai Chen, Yih-Giun Cherng, and Ying-Hsuan Tai. Yu-Hsiang Chung and Jhih-Wei Wang contributed equally to this work and should be regarded as co–first authors. All authors have read and agreed to the final version of the manuscript.

Funding

This study was supported by the funds of the Thomas and Dorothy MJ Toung Professorship in Anesthesiology.

Data availability

The data presented in this study are available on request from the corresponding author. The data are not publicly available due to the regulations of the Institutional Review Board.

Declarations

Ethics approval and consent to participate

We obtained the approval from the Taipei Medical University Joint Institutional Review Board (TMU-JIRB number: N202411041). The need for informed consent was waived by the Institutional Review Board due to the retrospective nature of this research.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.