The effect on cardiopulmonary function after thoracoplasty in pectus carinatum: a systematic literature review

Abstract

Objectives

Creating an aesthetically appealing result using thoracoplasty, especially when correcting extensive deformities, but only causing low morbidity, is challenging. The frequency of thoracoplasties in cases of pectus carinatum (PC) has increased due to improved experience and modified surgical techniques, resulting in low morbidity and low complication rates. The indications for surgical treatment are still controversial and, in most cases, remain aesthetic or psychological rather than physiological. However, whether cardiopulmonary function changes after surgical repair remains a matter of controversy. We sought to investigate and shed light on published knowledge regarding this question.

Methods

We searched MEDLINE and PubMed databases, using various defined search phrases and inclusion criteria, to identify articles on pre- and postoperative cardiopulmonary evaluation and outcomes.

Results

Six studies met the inclusion criteria: 5 studies evaluated patients with PC for cardiopulmonary outcomes after chest wall surgery and 1 did so following conservative compression treatment. In these studies, surgical and conservative correction of PC did not reduce absolute lung volumes and spirometric measurements and consequently had no pathogenic effect on cardiopulmonary function.

Conclusions

The results of this systematic review suggest that surgical correction of PC has no symptomatic pathogenic effect on cardiopulmonary function. The results, however, revealed both heterogeneity in the examinations used and inconsistent methods within each study. Further prospective trials with a stronger methodological design are necessary to objectively confirm that surgical correction of PC does not impair cardiopulmonary function.

Introduction

Pectus carinatum (PC), or keel chest or pigeon breast, is defined as an anterior protrusion of the sternum and adjacent costal cartilage and is the second most common congenital chest wall deformity [1]. There are 2 distinct types of PC: The condrogladiolar type presents as a protrusion of the lower two-thirds of the corpus sterni, whereas the chondromanubrial type is a protrusion of the superior part and the manubrium of the sternum [1]. Mixed forms with a protrusion of the manubrium and depression of the corpus, or vice versa, are called pectus arcuatum (PA) [2]. PC occurs with a male predominant gender ratio of 4:1 [3]. The overall incidence varies but is estimated at 1:1000 to 1:10 000 [3]. PC is reported to be 7 times less frequent than pectus excavatum (PE) [3]. Even though the exact cause for PC remains unknown, it is generally accepted that excessive growth of rib cartilage would cause the protruding sternum [4]. PC usually manifests during a growth spurt in the early teenage years [1].

Typical patient complaints include aesthetic concerns related to a disturbed body image and reduced quality of life [5]. In some patients, physical symptoms, such as respiratory symptoms, palpitations and skeletal discomfort, may be present [1]. Controversy remains, however, regarding the association between PC and respiratory dysfunction, because reported physiological studies on these patients are sparse [1, 5]. Especially compared to patients with PE, those with PC infrequently suffer from cardiorespiratory symptoms, so the question arises whether the correction of PC has any harmful effect on cardiopulmonary function [6].

Sparse data have been published on the indications for PC correction [2, 7]. The choice of treatment depends on the patient’s clinical presentation and on the surgeon’s preference and experience. In most cases, the indications for surgical treatment are aesthetic concerns but can also be based on accompanying psychological changes, especially in puberty and adolescence [2]. Beginning in the 1940s with open repairs, following Ravitch’s pioneering work [8], several modifications for the correction of PC have been published [3, 4, 6, 9, 10]. The traditional surgical repair described by Ravitch includes elevation of the pectoralis major and rectus abdominis muscles away from the sternum and adjacent ribs and resection of the deformed costal cartilages, including 1 or several sternum osteotomies, with or without strut placement [6]. Adapting the Nuss procedure for PC, Abramson et al. [11] established a minimally invasive technique. A thorough evaluation of postoperative complications of PC correction has recently been published [12].

Non-operative procedures for PC correction using a dynamic compression system have also been published [13, 14]. Guidelines [7] and reviews [1, 15, 16] have tried to summarize an ideal treatment plan, but no general consensus on how best to treat a PC deformity has been established.

PC corrections are being performed more and more often, but to date, there remains a lack of scientific evidence regarding any effects surgery has on cardiopulmonary function. This article points out many difficulties that we have with paediatric surgical problems and with identifying physiological changes after interventions. We presenta literature review focused on thephysiological changes accompanying treatment (operative or non-operative) of PC. We provide a systematic review of the available literature concerning cardiopulmonary function pre- and postoperatively, focusing on the highest quality evidence available, and determine the effect on cardiopulmonary function after correction of PC deformities to support clinical decision-making for treating physicians.

Materials and Methods

Literature search strategy

We completed a systematic literature search in May 2017, of publications in PubMed and MEDLINE databases. The keywords used for the search are shown in Fig. 1.

Figure 1. Keywords used for the search (PubMed/MEDLINE) and articles found (n = 1427).

Reference lists of relevant articles were reviewed manually to capture other potentially relevant publications. Two independent reviewers (S.S. and C.H.) examined potentially relevant articles using the following inclusion and exclusion criteria. No discrepancies occurred between the 2 reviewers while performing this review.

Selection of articles

To avoid the selection bias, specific inclusion and exclusion criteria were established before searching (Fig. 2). The selected articles were also evaluated regarding their level of evidence following the level of evidence pyramid [17]. All included studies were searched for the number of patients with PC, mean age (including range), pre- and postoperative diagnostics, surgical technique and statistical significant changes in any of the collected data.

Figure 2. Schematic search strategy of included and excluded articles.

Inclusion criteria

Inclusion criteria were surgical or conservative treatment for PC; preoperative evaluation of cardiac and/or pulmonary function and postoperative evaluation other than aesthetic outcome. No time frame was set.

Exclusion criteria

Exclusion criteria were postoperative evaluation based only onthe aesthetic outcome and patient satisfaction regarding aesthetics. Ther were no language restrictions.

Results

A total of 1427 articles were initially identified using the search terms. Six articles that fulfilled our search and exclusion criteria were selected for review (Fig. 2 and Table 1). The level of evidence was III in 5 studies [18–22] and IV in 1 study [23]. No randomized controlled trials were included since none was found. No studies showed a statistically significant result regarding postoperative cardiopulmonary function for patients with PC.

Table 1. Studies included in this review.

	Preoperative											Postoperative
First author, year of publication and reference	Total number of patients with PC and PE	Number of patients with PC	Spirometric measurements	Nitrogen washout	Cycle ergometer	Electrocardiogram	Echocardiography	Computed tomography	Chest radiograph	Pulmonary function		Number of patients with PC	Spirometric measurements	Nitrogen washout	Cycle ergometer	Electrocardiogram	Echocardiography	Computed tomography	Chest radiograph	Clinical evaluation	Questionnaire (email)	Questionnaire (telephone)	Results regarding pulmonary function (compared pre- and postoperatively)
Surgical treatment
Cahill et al. (1984) [19]	19	5	+	+	+	+	-	-	-	-		5	+	+	+	+	-	-	-	-	-	-	No significant difference
Derveaux et al.(1989) [22]	88	13	+	-	-	-	-	-	+	7		+	-	-	-	-	-	-	+	-	-	-	Reduced FEV1 and VC using percent predicted and no worsening of absolute values
Jaroszewski et al.(2007) [23]	320	41	+	-	-	+	+	+	+	+		41	-	-	-	-	-	-	-	+	+	+	Improved load-bearing capacity less dyspnoean and reduced pain
Oncel et al.(2013) [20]	77	31	-	-	-	+	+	+	+	+		?	-	-	-	-	-	-	-	-	-	+	Improved load-bearing capacity and lessdyspnoea
Bagheri et al.(2015) [18]	33	13	+	-	-	+	+	+	+	-		13	+	-	-	-	-	-	-	+	-	-	No significant difference
Conservative treatment Ates et al.(2013) [21]	53	53	+	-	-	-	+	+	-	-		48	+	-	-	-	+	+	-	-	-	-	No significant difference

FEV₁: forced expiratory volume in 1 _s; PC: pectus carinatum; PE: pectus excavatum; VC: vital capacity.

In the included studies (Table 1, number of patients with PC, preoperative), 156 patients underwent preinterventional diagnostics regarding cardiopulmonary parameters. The exact number of patients evaluated after PC correction in the included studies (Table 1, number of patients with PC, postoperative) could not be determined, because in 1 study no difference between PC and PE was made [20]. The exact demographics of the included patients could also not be evaluated, because the results were not divided into PC and PE patients in all studies (Table 1).

Selected studies

Cahill et al. [19] performed preoperative exercise and static pulmonary function tests on 19 patients with PE (n = 14) and PC (n = 5), including standard spirometric measurements of vital capacity (VC), forced expiratory volume and maximum voluntary ventilation. Lung volumes were calculated using a nitrogen washout method for measurement of functional residual capacity. A cycle ergometer was used to evaluate maximal work capacity. Additional tests included electrocardiogram, partial pressures (oxygen, carbon dioxide and nitrogen), oxygen consumption and total work time. The mean age at time of surgery was 13.8 (range 13-15) years. The same measurements were repeated 3-9 months after surgery. No specific surgical technique was mentioned.

All acquired postoperative data for the 5 patients with PC were within normal limits. No significant statistical changes could be measured in the PC patient group regarding any collected data.

Derveaux et al. [22] included 88 patients with PE (n =54), PC (n = 13), pectus arcuatum (PA) (n =13) and PE with additional severe scoliosis (n = 8), for whom radiology and lung function evaluations were available. The mean age of patients with PC at the time of surgery was 16 (range 13-19) years. Preoperative cardiopulmonary evaluation consisted of forced expiratory volume in 1 s (FEV₁) and VC. The same measurements were repeated following surgery for only 7 patients with PC, but the exact diagnostic method of these measurements was not mentioned. Follow-up time was 8 (range 1-20) years, but the exact time of the postoperative evaluation of lung function was not mentioned either. Surgical correction was performed using a variant of the Baronofsky technique [24], which is a minor variant of the Ravitch repair [8].

Their data showed that there was no decrease in the absolute values for FEV₁ and VC. Significant worsening, when expressed in percent predicted for children by Zapletal et al. [25], was shown.

This finding was interpreted to be a smaller increase in lung volumes during adolescence in patients with PC and PE than in healthy subjects due to extensive cartilage resection resulting in decreased chest wall compliance. This conclusion, however, was only discussed; no objective proof was presented.

Jaroszewski et al. [23] evaluated 21 years of experience correcting pectus chest deformities in adults. Three-hundred twenty patients with PE (n = 268), PC (n = 41) and PA (n = 11) were included in this study. The mean age at the time of surgery was 27 (range 19-67) years. Preoperative cardiopulmonary and structural diagnostics included chest radiograph or computed tomography (CT) scan, electrocardiogram and echocardiography, pulmonary function tests and spirometry. All patients underwent variations of a highly modified Ravitch repair using a sternal strut. Follow-up examinations after strut removal were performed after 26 months (range 1 month-21 years) at office visits or using telephone or email questionnaires. No postoperative cardiopulmonary diagnostics were conducted for patients following surgical repair.

Therefore, no objective data are available for patients with PC. Moderate to marked improvement in exercise tolerance with much less dyspnoea and increased endurance was reported within 4 months after PE or PC repair by all but 6 patients, 3 of whom had repairs of recurrent deformities. Preoperative chest pain was resolved, or considerably improved, in all but 2 patients after repair.

Oncel et al. [20] included 77 patients with PE (n = 46) and PC (n = 31). The mean age at the time of surgery was 17 (range 10-22) years. Preoperative cardiopulmonary and structural diagnostics consisted of chest radiograph, CT scan of the chest, electrocardiogram, echocardiogram and respiratory function tests. All patients with PC underwent a Ravitch repair [8]. Postoperative follow-up examinations were conducted by telephone questionnaires.

Three months to 3 years after surgery, all patients were contacted by telephone. Fifteen could not be located and were lost to follow-up. The remaining 62 patients reported marked improvement in exercise tolerance with much less dyspnoea. All but 3 patients, who had only limited exercise ability before the repair, showed good physical activity 6 months after the operation. It is not stated how many patients with PC were evaluated. No postoperative diagnostics were conducted for patients following surgical repair, so no objective data were available for patients with PC.

Bagheriet al. [18] included 33 patients with PE (n = 20) and PC (n = 13). The mean age at the time of surgery was 15 (range 1022) years. Preoperative evaluation consisted of chest radiograph, chest CT scan, echocardiography, pulmonary function tests measuring FEV₁, functional residual capacity and inspiratory VC. The same diagnostic tests were performed over the course of 1 year after the Ravitch repair surgery. Postoperative pulmonary function in patients with PC was not significantly different from that before the operation.

Ates et al. [21] included 53 patients with PC. The mean age at the beginning of the non-operative treatment using compression was 12 (range 5-18) years. Pretreatment evaluation consisted of a chest CT scan, echocardiography and pulmonary function testing (forced VC and FEV₁). The average treatment time was 21.3 months. The same diagnostics were repeated after bracing therapy. Forced VC and FEV1 were correlated with the predicted values for age according to Zapletal et al. [25]. A statistical workup was performed. The average follow-up period was 24.1 months. No statistically significant change in pulmonary function was detected.

Discussion

When looking at a patient with a PC deformity, evaluating CT scans or radiographs and comparing them with similar scans of a patient with a PE deformity or even a normal chest scan, it seems obvious that, when performing a surgical or even a conservative correction, the intrathoracic volume will decrease and thus space for the heart and lungs will be reduced as well. In contrast to repair of PE, which has known benefits from expanding the space in the lower mediastinum for improved cardiac (particularly the right heart) function, the benefit of repair of PC is much less obvious and harder to understand. Because most patients with PC do not suffer from any clinical or physiological symptoms but rather from psychological limitations, the question arises whether the reduction of intrathoracic volume also reduces cardiopulmonary function and, consequently, lessens the patient’s cardiopulmonary strain capacity.

Prior to any decision and intervention, most institutions perform a thorough evaluation of the patient, using spirometric measurements, electrocardiogram, echocardiography and radiographs (Table 1), but only 4 papers reported an objective, reproducible postinterventional cardiopulmonary evaluation [16, 17, 20, 21]. The work of Ates et al. [21] was included in this literature research because of their thorough postoperative evaluations. No significant reduction of spirometric measurements was found. However, the surgically treated patients cannot be included in this statement because the scarring around the sternum following resection of cartilage must not be disregarded. Twenty-five patients were evaluated for cardiopulmonary factors (e.g. spirometric measurements and cycle ergometer; Table 1) after the surgical correction of PC [18, 23]. The follow-up examinations performed in the other included papers comprised only a clinical evaluation and, in most cases, were limited to a retrospective questionnaire conducted by phone or email, thus providing subjective data. The exact kind of questionnaire was not mentioned in the reported studies [18, 19].

All 25 patients with PC mentioned in the included studies who were evaluated after the operation underwent the original Ravitch repair or modifications. No data were available concerning the Abramson et al. [11] repair using a steel strut for ventral compression. Depending upon the surgical method of thoracoplasty, most patients had PC repair using variations of the Ravitch technique. Only 1 paper reported a different surgical technique but did no cardiopulmonary follow-up examination. So, we were unable to compare cardiopulmonary function in patients undergoing different surgical techniques. Derveaux et al. [22] reported a statistically significant worsening of pulmonary function, although only with regard to the percent predicted for children. In this case, it is noteworthy that an extensive resection, even in adolescence, causes a mild restriction of the growing thorax. The absolute measurements, however, showed no significant reduction. This observation might mean that the predictions used by Zapletal et al. [25] cannot be used in this specific population because the patients were already 16 years of age at the time of surgery or that the extensive resection of cartilage and the scarring that follows restrict thoracic growth, resulting in lung volumes that are smaller than predicted. The patients in this study,however, reported no decrease in load-bearing capacity or any other limitations in their daily life.

When we evaluated the available data, the results indicated that there was no relevant unfavourable decrease in cardiopulmonary function, that the patients were content with the outcome concerning physical capability and that most would undergo surgical correction again. In most cases, however, there were no objective data, and all investigations were retrospective. This systematic review also demonstrates a wide variety of preoperative cardiopulmonary function tests and radiological examinations. However, postoperative measurements, if reported, were inconsistent across the available studies. Furthermore, the majority of these studies did not address postoperative measurements for patients with PC. Specifically, postoperative standardized cardiopulmonary function tests were not described in detail. Few researchers performed pre- and postoperative measurements. Quality evidence of results was available only for the non-surgical treatments. So, this review also highlights several research needs: in particular, the lack of rigorous data concerning postoperative standardized cardiopulmonary function.

To further strengthen these results, more stringent data are necessary. There is a need for prospectively gathered data of relevant measures about both physiology and perceived benefit from that correction of PC. An improved understanding of the health needs of patients with PC is required, because the impact of PC deformities (measured by instruments designed to evaluate health-related quality of life) is evaluated infrequentlyin patients with PC.

Limitations

The strengths of this review include a comprehensive search strategy and robust methods for study selection. Limitations of this analysis include the small number and quality of publications available. Another important limitation is that all reviewed papers were case studies, and none offered direct comparisons in the form of randomized controlled trials. Multiple examinations (function tests and/or radiological examinations) were studied. However, they lacked a consistent standard of pre- and postoperative measurements on cardiopulmonary function in the treatment of PC deformities, which prevented conclusive comparisons between the studies and their individual results.

Conclusion

The results of this systematic review revealed that the studies performed heterogeneous examinations within each category and were inconsistent in the methods used to determine cardiopulmonary function outcome. Furthermore, in the majority of studies, the evidence level was low to moderate.

As highlighted in this review, further investigation with higher level evidence and methodological rigour is necessary to determine any beneficial or adverse effect of surgical treatment of PC on cardiopulmonary function.

For that reason, a prospective study began in 2013 in our Department of Plastic, Reconstructive andAesthetic Surgery, Innsbruck Medical University. It was approved by the local ethics committee and sponsored by the Austrian Science Fund (FWF) [KLI312] and is registered at clinicaltrials.gov (NCT02163265). Patients with PC are being thoroughly evaluated in terms of cardiopulmonary function and quality of life with standardized measurements pre- and postoperativelyafter surgical correction of PC.