Comparison of 30-Day Outcomes Between Thoracoscopic and Open Lobectomy for Congenital Pulmonary Lesions

Abstract

Objective:

To compare postoperative length of stay and 30-day outcomes between thoracoscopic and open lobectomy performed on a nonemergency basis for congenital pulmonary lesions using a validated national database.

Materials and Methods:

We identified all nonemergency lobectomies performed on patients with congenital pulmonary lesions in the 2012 National Surgical Quality Improvement Program (NSQIP) Pediatric database and compared demographic, clinical, and 30-day outcome characteristics between patients who underwent an open or thoracoscopic lobectomy. Logistic regression with Firth's penalized likelihood bias-reduction method was used to determine predictive risk factors for a postoperative length of stay (LOS) of >3 days.

Results:

Of 101 patients included, 40 (39%) underwent thoracoscopic lobectomy. In comparison with patients undergoing thoracoscopic lobectomy, patients undergoing open lobectomy were significantly more likely to be admitted prior to surgery, be American Society of Anesthesiologists Class ≥3, receive oxygen support prior to surgery, and have other congenital anomalies or cardiac risk factors. Both groups had similar total operative times (open versus thoracoscopic, 150 versus 173 minutes; P=.216). Patients undergoing open lobectomy had longer postoperative LOS (4 versus 3 days; P=.001) and more often received an intraoperative or postoperative transfusion (12% versus 0%; P=.003). The procedure type was not an independent risk factor for postoperative LOS >3 days in the multivariable analysis.

Conclusions:

Patients undergoing thoracoscopic lobectomy have fewer comorbidities at baseline, receive fewer perioperative transfusions, and have a shorter postoperative LOS. Accrual of additional patients within the NSQIP Pediatric database will allow for further risk-adjusted analyses to control for differences in baseline characteristics between patients undergoing open and thoracoscopic resections.

Introduction

Congenital pulmonary lesions refer to a class of lung lesions including congenital pulmonary airway malformation, congenital lobar overinflation, bronchopulmonary sequestration, and bronchogenic cysts.¹ These rare anomalies of the lung are frequently discovered during routine prenatal ultrasound with a reported incidence of 1 in 12,000 live births, a value that may be increasing in recent years because of improved prenatal ultrasound screening.^2–4 Patients diagnosed with a congenital pulmonary lesion are often treated with a lobectomy because of the risk of infection, spontaneous pneumothorax, or malignant degeneration.^2,5–8 Some patients present with severe respiratory and cardiac symptoms due to these malformations and must undergo urgent surgical intervention. However, in most cases, lobectomy can be performed on an elective basis, as asymptomatic patients can be safely observed for weeks or months after birth, when anesthesia is safer^9,10 and the operation is technically more straightforward.¹¹

Lobectomy has traditionally been performed via open thoracotomy, but at multiple tertiary pediatric centers, nonemergency lobectomies are now being performed using a thoracoscopic approach. Potential advantages of a thoracoscopic approach include improved postoperative pain control, decreased length of stay (LOS), and improved cosmesis.^12–17 In addition, previously reported potential musculoskeletal sequelae of performing open thoracotomies in young children, including winged scapula and atrophy of the serratus anterior muscle, scoliosis, and pectus excavatum, may be avoided.^15,18–20 Potential disadvantages of a thoracoscopic approach include longer operative times, prolonged exposure to general anesthesia, and greater technical complexity.^13,21,22

Adult studies support a lower incidence of postoperative complications and a shorter LOS in patients undergoing thoracoscopic lobectomy compared with open lobectomy.²³ However, reports investigating thoracoscopic pulmonary lobectomy in children have been limited to single-institutional studies because of the rarity of congenital pulmonary lesions. The purpose of this study was to compare postoperative LOS and 30-day outcomes between thoracoscopic and open lobectomies performed on a nonemergency basis for congenital pulmonary lesions using a validated multi-institutional national database.

Materials and Methods

The National Surgical Quality Improvement Project (NSQIP) Pediatric is a multispecialty program managed by the American College of Surgery in collaboration with the American Pediatric Surgical Association. The program reports peer-reviewed, risk-adjusted 30-day postoperative outcomes on systematically sampled surgical cases performed on patients under 18 years of age from participating institutions for the purpose of benchmarking and quality improvement. Data are collected by a dedicated trained surgical clinical reviewer. Included cases are selected based on Current Procedural Terminology (CPT) codes using the NSQIP 8-day cycle-based systematic sampling of 35 procedures per cycle. At least 124 variables are collected from the medical records and the patients and their families including information on demographics, surgical profile, preoperative and intraoperative variables, and postoperative occurrence and discharge variables.^24–26 Surgical cases for acute traumatic injuries are not included in the database.²⁷

All pulmonary lobectomies performed on patients <18 years old with congenital pulmonary lesions in the 2012 NSQIP Pediatric database were identified using CPT codes and International Classification of Diseases, Ninth Revision, Clinical Modifications (ICD-9-CM) diagnostic codes. CPT codes reflect the final procedure performed and not the scheduled procedure. NSQIP Pediatric does not publicly report information on the initially scheduled procedure; therefore, instances of conversion of a minimally invasive procedure to an open procedure cannot be determined. Patients were defined as undergoing an open lobectomy if they had a CPT code of 32480 or 32482 or thoracoscopic lobectomy if they had a CPT code of 32663 or 32670. Patients with congenital pulmonary lesions were defined as all patients with ICD-9-CM code 492.8 (other emphysema), 748.4 (congenital cystic lung), 748.5 (agenesis, hypoplasia, and dysplasia of lung), 748.60 (unspecified anomaly of lung), 748.69 (other congenital anomalies of lung), or 770.2 (interstitial emphysema and related conditions). Patients who reported a current or prior history of cancer at the time of their surgery or who underwent an emergency operation were excluded.

We compared demographic, clinical, and 30-day outcome characteristics between patients who underwent an open or thoracoscopic lobectomy. Neonates were defined as all patients ≤30 days old at the time of surgery, and prematurity was defined as all patients born ≤37 weeks of gestational age. Minor complications were defined as superficial surgical-site infections and urinary tract infections within 30 days. Major complications were defined as deep or organ-space surgical-site infections, unplanned re-intubation, pneumonia, ventilator dependence for >24 hours, sepsis, septic shock, intraoperative or postoperative blood product transfusion within 72 hours of surgery, unplanned reoperation, pulmonary embolism, progressive renal insufficiency, acute renal failure, cerebrovascular accident, coma for >24 hours, cardiac arrest, myocardial infarction, venous thromboembolism, or death within 30 days.

Continuous variables were compared using Wilcoxon rank-sum tests, and categorical variables were compared using Pearson chi-square tests or Fisher's exact tests where appropriate. P values<.05 were considered statistically significant. Because of the sparseness of the data and quasicomplete separation of the data for some variables, Firth's penalized likelihood bias-reduction method²⁸ was used to determine the risk factors for a postoperative LOS >3 days. Three postoperative days is the expected length of stay for patients undergoing lobectomy for a congenital pulmonary lesion based on our clinical experience and existing literature.^{11,17,29–31} All factors with P<.1 after univariable analysis were considered for inclusion in the multivariable model, which was fit using backward stepwise regression. All analyses were performed using SAS version 9.3 software (SAS Institute, Cary, NC).

Results

Of the 101 patients who underwent a nonemergency lobectomy for a congenital pulmonary lesion, 40 (39%) underwent thoracoscopic lobectomy. Table 1 describes demographics and preoperative characteristics for patients who underwent open or thoracoscopic lobectomy. Patients undergoing an open lobectomy were less likely to have surgery during the neonatal period and were more likely to have preexisting cardiac risk factors and congenital malformations other than the congenital pulmonary lesion. They were also more likely to require preoperative oxygen support, have an American Society of Anesthesiologists (ASA) Class ≥3, and be admitted prior to the day of surgery.

Table 1.

Demographics and Preoperative Characteristics of Patients Undergoing Open or Thoracoscopic Pulmonary Lobectomy for Congenital Pulmonary Lesions

Characteristic	Open (n=61)	Thoracoscopic (n=40)	P value
Patient demographics
Age at surgery (months)^a	5 (2–9)	6 (4–10)	.123
Neonate (age ≤30 days at surgery)	10 (16)	11 (28)	.046^b
Prematurity (birth prior to 37 weeks of gestation)	12 (20)	3 (8)	.095
Female	24 (39)	16 (40)	.948
Race			.976
White	41 (67)	27 (68)
Nonwhite	20 (32)	13 (33)
Hospital admission
Transfer status			.046^b
From home, clinic, doctor's office	50 (82)	39 (98)
From emergency department	1 (2)	0 (0)
Other	10 (16)	1 (3)
Admitted 1 or more days prior to date of operation^c	12 (20)	1 (3)	.013^b
Comorbidities/complications
Weight loss >10%/failure to thrive	3 (5)	1 (3)	1.000
Preexisting cardiac risk factors	16 (26)	2 (5)	.007^b
Preoperative ventilator dependence	3 (5)	0 (0)	.275
Preoperative oxygen support	8 (13)	0 (0)	.021^b
Other congenital malformation	31 (51)	12 (30)	.039^b
Small for gestational age	0 (0)	0 (0)	NA
Bleeding or hematologic disorder	0 (0)	0 (0)	NA
ASA Class ≥3	28 (46)	6 (15)	.001^b
Blood transfusions within 48 hours before surgery	0 (0)	0 (0)	NA

Data are frequency (percentage) for categorical variables.

Data are median (interquartile range) for continuous variables.

Significant difference at P<.05.

Data are missing for 1 patient.

ASA, American Society of Anesthesiologists; NA, not applicable.

Operative details and 30-day postoperative outcomes of patients who underwent open and thoracoscopic lobectomy are described in Table 2. There were no differences in total operative time, duration of time in the operating room, or time under general anesthesia. Patients undergoing thoracoscopic procedures had shorter postoperative LOS and were less likely to receive an intraoperative transfusion or a transfusion within 72 hours of surgery. There were no differences in other postoperative complication rates.

Table 2.

Postoperative 30-Day Outcomes of Patients Undergoing Open or Thoracoscopic Pulmonary Lobectomy for Congenital Pulmonary Lesions

Outcome	Open (n=61)	Thoracoscopic (n=40)	P value
Duration of time in the operating room (minutes)	252 (197–299)	271 (160–340)	.413
Anesthesia time (minutes)^a	262 (208–316)	283 (181–343)	.604
Total operation time (minutes)	150 (102–188)	173 (109–235)	.216
Postoperative length of stay (days)^b	4 (3–7)	3 (2–4)	.001^c
Any minor complication	1 (2)	1 (3)	1.000
Superficial incisional surgical-site infection	0 (0)	1 (3)	.392
Urinary tract infection	1 (2)	0 (0)	1.000
Any major complication	16 (26)	4 (10)	.072
Deep or organ-space surgical-site infection	0 (0)	1 (3)	.396
Unplanned re-intubation	1 (2)	0 (0)	1.000
Pneumonia	0 (0)	1 (3)	.396
Ventilator dependence for >24 hours^d	6 (10)	1 (3)	.234
Sepsis or septic shock	0 (0)	1 (3)	.396
Unplanned return to operating room	1 (2)	1 (3)	1.000
Intraoperative transfusion or transfusion within 72 hours	12 (19)	0 (0)	.003^c
Still in the hospital after 30 days^b	1 (2)	1 (3)	1.000
Unplanned re-admission	2 (3)	0 (0)	.519

Results are expressed as frequency (percentage) for categorical variables and median (interquartile range) for continuous variables. There were no incidents of the following major complications in either group: pulmonary embolism, progressive renal insufficiency, acute renal failure, cerebrovascular accidents, coma for >24 hours, cardiac arrest, myocardial infarction, venous thromboembolism, or death.

Time during which patient is under care of an anesthesiologist, including during postoperative recovery.

Data are missing for 1 patient.

Significant difference at P<.05.

Data are missing for 4 patients.

Table 3 presents results of a logistic regression model used to assess variables associated with a postoperative LOS >3 days. One patient was excluded from this analysis because data on LOS were not available. After adjustment, the surgical approach (open versus thoracoscopic) was no longer a significant predictor of LOS >3 days. However, ASA Class≥3 was found to be a significant predictor of LOS.

Table 3.

Firth's Penalized Logistic Regression Models for Patients Undergoing Pulmonary Lobectomy for Congenital Pulmonary Lesions Who Had a Postoperative Length of Stay of >3 Days

	Unadjusted			Adjusted
Risk factor	OR	95% CI	P value	OR	95% CI	P value
Open procedure	3.40	1.45, 7.93	.005^a	2.38	0.96, 5.90	.060
ASA Class ≥3	5.68	2.23, 14.47	<.001^a	4.47	1.70, 11.76	.002^a
Cardiac risk factors	1.82	0.81, 4.09	.147
Neonate (age ≤30 days)	32.20	1.62, 638.33	.023^a
Transfer status (emergency room or other)	32.20	1.62, 638.33	.023^a
Oxygen support	18.98	0.87, 416.22	.062
Weight loss >10%/failure to thrive	10.61	0.40, 285.04	.159

Significant difference at P<.05.

ASA, American Society of Anesthesiologists; CI, confidence interval; OR, odds ratio.

Discussion

This study represents the largest multi-institutional compilation to date of patients undergoing nonemergency lobectomy for congenital pulmonary lesions using validated data with standardized definitions of postoperative outcomes. The data suggest that patients undergoing thoracoscopic lobectomy have fewer comorbidities at baseline and receive fewer perioperative transfusions. Although initial analysis suggested that patients undergoing thoracoscopic lobectomy have shorter postoperative LOS, the thoracoscopic approach was no longer a significant predictor of a shorter postoperative LOS in our multivariable risk adjusted analysis (P=.06).

Existing single-institution studies report that patients with more comorbidities and symptomatic pulmonary disease are more likely to undergo open thoracotomy. Our study also supports that patients undergoing open lobectomy are more likely to be categorized as ASA Class ≥3, to have preoperative cardiac risk factors, and to require preoperative oxygen support. Several surgical groups have described delaying thoracoscopic lobectomy when possible to allow the patient to grow, reducing the risks of general anesthesia exposure at a young age⁹ and decreasing the surgical difficulty of thoracoscopic lobectomy.^2,32,33 However, one study proposes that thoracoscopic lobectomy is safe for infants <10 kg and may avoid inflammatory changes associated with delayed treatment.²² Our data demonstrate that the thoracoscopic approach is prevalently used in neonates and infants and suggest that as surgeons have become more proficient with thoracoscopic lobectomies, they are now performing them in younger patients.

One potential disadvantage of thoracoscopic lobectomy frequently mentioned is that operative times may be longer and patients may be unnecessarily exposed to prolonged general anesthesia. Several studies have suggested that thoracoscopic operative times decrease as institutional experience and surgeon volume increase.^29,34 Our study demonstrated no difference in operative or anesthetic times for patients undergoing open and thoracoscopic lobectomy for congenital pulmonary lesions. This finding is consistent with a recent meta-analysis¹³ of six retrospective cohort studies, which reported similar operatives in patients undergoing open and thoracoscopic lobectomies.

In our study, patients undergoing an open lobectomy were more likely to require an intraoperative or postoperative blood transfusion than patients undergoing a thoracoscopic lobectomy. One previous pediatric study suggested the opposite,²⁹ but our results are consistent with adult literature on thoracoscopic lobectomies performed for other etiologies.^35–37 Although there is a concern over the inability to control bleeding should it occur during a thoracoscopic surgery, there was no morbidity associated with extensive blood loss among the patients undergoing thoracoscopic lobectomy who required transfusion in our current study.

We recognize several limitations in the current study. Although this is the largest sample size to date of patients with congenital pulmonary lesions, the sample size was still too small to perform additional multivariable analyses, including examining the need for a postoperative blood transfusion or other complications. Additional accrual of data into the NSQIP Pediatric database and participation by additional institutions will allow for future analysis with increased power to test these associations. Also, because NSQIP Pediatric data are deidentified and coding reflects the final procedure performed, we are unable to assess whether any patients in our dataset underwent a planned thoracoscopic lobectomy that was intraoperatively converted to open lobectomy. Currently, NSQIP Pediatric documents but does not publicly report whether patients in the dataset underwent minimally invasive procedures that were converted to open procedures. Therefore, using publicly reported data, we have no means of identifying those patients who were converted from a thoracoscopic procedure to an open procedure, but we do know that they would be part of the open procedure cohort of patients. Finally, we do not have data on pain control, duration of thoracostomy tube placement, or long-term morbidities associated with thoracotomy in children, or complications beyond 30 days such as scoliosis or chest wall deformity.

The current study represents the largest compilation of patients undergoing nonemergency lobectomy for congenital pulmonary lesions using standardized and validated clinical data collection with defined postoperative outcomes. Accrual of additional patients within the NSQIP Pediatric database will allow for further risk-adjusted analyses of outcomes to control for differences in baseline characteristics between patients undergoing open and thoracoscopic resections. Future research should include evaluation of postoperative pain control and duration of thoracostomy tube placement using other databases.

Conclusions

Patients undergoing thoracoscopic lobectomy do not experience longer operative times, but they have a shorter postoperative LOS and receive fewer perioperative transfusions. The type of surgical approach selected, however, is not predictive of discharge >3 days after operation.

Footnotes

Acknowledgments

This project is supported by intramural funding from the Center for Surgical Outcomes Research and the Department of Pediatric Surgery at Nationwide Children's Hospital.

Disclosure Statement

No competing financial interests exist.

References

Puligandla

, Laberge

. Congenital lung lesions. Clin Perinatol, 2012; 39:331–347.

Laberge

J-M

, Puligandla

, Flageole

. Asymptomatic congenital lung malformations. Semin Pediatr Surg, 2005; 14:16–33.

Nasr

, Himidan

, Pastor

, Taylor

, Kim

. Is congenital cystic adenomatoid malformation a premalignant lesion for pleuropulmonary blastoma?. J Pediatr Surg, 2010; 45:1086–1089.

Duncombe

, Dickinson

, Kikiros

. Prenatal diagnosis and management of congenital cystic adenomatoid malformation of the lung. Am J Obstet Gynecol, 2002; 187:950–954.

Aziz

, Langer

, Tuuha

, Ryan

, Ein

, Kim

. Perinatally diagnosed asymptomatic congenital cystic adenomatoid malformation: To resect or not?. J Pediatr Surg, 2004; 39:329–334.

Coran

, Drongowski

. Congenital cystic disease of the tracheobronchial tree in infants and children: Wxperience with 44 consecutive cases. Arch Surg, 1994; 129:521–527.

Conforti

, Aloi

, Trucchi

, et al. Asymptomatic congenital cystic adenomatoid malformation of the lung: Is it time to operate?. J Thorac Cardiovasc Surg, 2009; 138:826–830.

Wong

, Vieten

, Singh

, Harvey

, Holland

. Long-term outcome of asymptomatic patients with congenital cystic adenomatoid malformation. Pediatr Surg Int, 2009; 25:479–485.

Thomas

. Reducing the risk in neonatal anesthesia. Paediatr Anaesth, 2014; 24:106–113.

10.

Sanders

, Davidson

. Anesthetic‐induced neurotoxicity of the neonate: Time for clinical guidelines?. Paediatr Anaesth, 2009; 19:1141–1146.

11.

Chiu

, Flake

. Congenital lung lesions. In: Mattei

(ed). Fundamentals of Pediatric Surgery. New York: Springer, 2011, pp. 293–297.

12.

Koontz

, Oliva

, Gow

, Wulkan

. Video-assisted thoracoscopic surgical excision of cystic lung disease in children. J Pediatr Surg, 2005; 40:835–837.

13.

Nasr

, Bass

. Thoracoscopic vs open resection of congenital lung lesions: A meta-analysis. J Pediatr Surg, 2012; 47:857–861.

14.

Rahman

, Lakhoo

. Comparison between open and thoracoscopic resection of congenital lung lesions. J Pediatr Surg, 2009; 44:333–336.

15.

, Farmer

, Nobuhara

, Miniati

, Lee

. Thoracoscopic versus open resection for congenital cystic adenomatoid malformations of the lung. J Pediatr Surg, 2008; 43:35–39.

16.

Diamond

, Herrera

, Langer

, Kim

. Thoracoscopic versus open resection of congenital lung lesions: A case-matched study. J Pediatr Surg, 2007; 42:1057–1061.

17.

Albanese

, Rothenberg

. Experience with 144 consecutive pediatric thoracoscopic lobectomies. J Laparoendosc Adv Surg Tech A, 2007; 17:339–341.

18.

Jaureguizar

, Vazquez

, Murcia

, Diez Pardo

. Morbid musculoskeletal sequelae of thoracotomy for tracheoesophageal fistula. J Pediatr Surg, 1985; 20:511–514.

19.

Westfelt

, Nordwall

. Thoracotomy and scoliosis. Spine, 1991; 16:1124–1125.

20.

Lawal

, Gosemann

J-H

, Kuebler

, Glüer

, Ure

. Thoracoscopy versus thoracotomy improves midterm musculoskeletal status and cosmesis in infants and children. Ann Thorac Surg, 2009; 87:224–228.

21.

Rothenberg

. First decade's experience with thoracoscopic lobectomy in infants and children. J Pediatr Surg, 2008; 43:40–44; discussion 45.

22.

Rothenberg

, Kuenzler

, Middlesworth

, et al. Thoracoscopic lobectomy in infants less than 10 kg with prenatally diagnosed cystic lung disease. J Laparoendosc Adv Surg Tech A, 2011; 21:181–184.

23.

Paul

, Sedrakyan

, Chiu

, et al. Outcomes after lobectomy using thoracoscopy vs thoracotomy: A comparative effectiveness analysis utilizing the Nationwide Inpatient Sample database. Eur J Cardiothorac Surg, 2013; 43:813–817.

24.

Dillon

, Hammermeister

, Morrato

, et al. Developing a NSQIP module to measure outcomes in children's surgical care: Opportunity and challenge. Semin Pediatr Surg, 2008; 17:131–140.

25.

Raval

, Dillon

, Bruny

, et al. Pediatric American College of Surgeons National Surgical Quality Improvement Program: Feasibility of a novel, prospective assessment of surgical outcomes. J Pediatr Surg, 2011; 46:115–121.

26.

Raval

, Dillon

, Bruny

, et al. American College of Surgeons National Surgical Quality Improvement Program Pediatric: A phase 1 report. J Am Coll Surg, 2011; 212:1–11.

27.

American College of Surgeons National Surgical Quality Improvement Program—Pediatrics. Available at http://pediatric.acsnsqip.org/index.jsp (accessed May 12, 2014).

28.

Firth

. Bias reduction of maximum likelihood estimates. Biometrika, 1993; 80:27–38.

29.

Kunisaki

, Powelson

, Haydar

, et al. Thoracoscopic vs open lobectomy in infants and young children with congenital lung malformations. J Am Coll Surg, 2014; 218:261–270.

30.

Tsai

, Liechty

, Hedrick

, et al. Outcomes after postnatal resection of prenatally diagnosed asymptomatic cystic lung lesions. J Pediatr Surg, 2008; 43:513–517.

31.

Azizkhan

, Crombleholme

. Congenital cystic lung disease: Contemporary antenatal and postnatal management. Pediatr Surg Int, 2008; 24:643–657.

32.

Samuel

, Burge

. Management of antenatally diagnosed pulmonary sequestration associated with congenital cystic adenomatoid malformation. Thorax, 1999; 54:701–706.

33.

Sauvat

, Michel

J-L

, Benachi

, Emond

, Revillon

. Management of asymptomatic neonatal cystic adenomatoid malformations. J Pediatr Surg, 2003; 38:548–552.

34.

Seong

, Kang

, Kim

J-T

, Moon

, Park

, Kim

. Video-assisted thoracoscopic lobectomy in children: Safety, efficacy, and risk factors for conversion to thoracotomy. Ann Thorac Surg, 2013; 95:1236–1242.

35.

Sugiura

, Morikawa

, Kaji

, Sasamura

, Kondo

, Katoh

. Long-term benefits for the quality of life after video-assisted thoracoscopic lobectomy in patients with lung cancer. Surg Laparosc Endosc Percutan Tech, 1999; 9:403–408.

36.

Demmy

, Curtis

. Minimally invasive lobectomy directed toward frail and high-risk patients: A case-control study. Ann Thorac Surg, 1999; 68:194–200.

37.

McKenna

Jr , Houck

, Fuller

. Video-assisted thoracic surgery lobectomy: Experience with 1,100 cases. Ann Thorac Surg, 2006; 81:421–426.