Current Practices in the Management of Congenital Diaphragmatic Hernia Patients Requiring Extracorporeal Membrane Oxygenation: Results of an International Survey of Pediatric Surgeons

Abstract

Introduction:

There is little consensus on optimal management for congenital diaphragmatic hernia extracorporeal membrane oxygenation (CDH ECMO) patients. Meaningful comparisons of the various approaches have been limited due to the low number of cases in institutions. In addition, the multidisciplinary reliance and rigid institutional framework of ECMO serve to further limit exposure to alternative practices. The goal of this study is to survey the international pediatric surgery community to describe the current practice trends.

Methods:

A survey was electronically distributed to the international pediatric surgical community. The results were evaluated using statistical analysis.

Results:

A total of 123 pediatric surgeons completed the survey, of whom 89% work at institutions offering both venoatrial (VA) and venovenous (VV) ECMO. Although 69% perform VA ECMO for CDH, only 46% felt VA was the “optimal method.” Among VV proponents, 21% believe the rate of VV to VA conversion to be <5% and 16% believe it to be >30% compared with 0% and 40% in VA proponents. Distribution of timing of repair: 46% post-ECMO repair, 22% early ECMO repair, 15% whenever stabilized on ECMO, and 14% late ECMO repair. Sixty-four percent (71/111) would perform an ECMO CDH repair in the unweanable patient and 27% (30/111) report successful decannulation after repair of a patient who was unweanable on ECMO for 2 weeks. Ninety-two percent do not perform exit-to-ECMO.

Conclusion:

There are significant practice variations in the management of CDH ECMO. Majority of pediatric surgeons perform VA ECMO in CDH patients; however, a significant percentage of those believe VV to be more optimal. This discrepancy is not accounted for by the VA-only institutions. Although post-ECMO CDH repair is the most common approach, the majority would perform a repair “on ECMO” if the patient was unweanable. In addition, although many pediatric surgeons believe the “last ditch repair” for the unweanable patient to be futile, 27% have reported success. Exit-to-ECMO for CDH remains a minority practice.

Introduction

Congenital diaphragmatic hernia (CDH) is a rare congenital anomaly that is commonly associated with varying degrees of lung hypoplasia and pulmonary hypertension, with the most severe forms often requiring extracorporeal membrane oxygenation (ECMO).¹ Despite continued advances in the field, there continues to be a wide variation in practice and little consensus regarding even the most basic elements of management, including mode of ECMO, timing of diaphragm repair in relation to ECMO, role of “last ditch” diaphragm repair for patients failing to improve, and the potential indications for exit-to-ECMO. The goal of this survey is to describe the practice patterns of the international pediatric surgery community with regard to the critical aspects of management of these most severe cases of CDH.

Materials and Methods

After approval by our institutional review board and the International Pediatric Endosurgery Group (IPEG) Research Committee, a link to a 10-question online survey created on Google Survey was sent by email to all registered members of IPEG. In addition, the link was posted to multiple international pediatric surgery Facebook groups, including Stay Current in Pediatric Surgery, Pediatric Surgery International Community, Syrian Pediatric Surgery Association, Pediatric Surgery Ludhiana, and Saudi Pediatric Surgery Club. Responses were collected anonymously from September to October 2016. The results were then analyzed using descriptive statistics, odds ratio (OR), and chi square test by a biostatistician.

Results

Demographics

The survey was completed by 123 pediatric surgeons. The majority of respondents were from North America (68.3%; n = 84), followed by Europe (13%; n = 16) (Table 1). The years of clinical practice varied widely (Table 2).

Table 1.

Geographic Locations of Respondents

Region	Respondents, n (%)
North America	84 (68.3)
Europe	16 (13)
South America	8 (6.5)
Asia	10 (8.1)
Africa	2 (1.6)
Oceania	3 (2.4)

Table 2.

Clinical Years in Practice

Years in practice	Respondents (%)
0–5	22.8
6–10	30.1
11–20	29.3
>20	17.9

Venoatrial versus venovenous ECMO

Ninety-one out of 123 (74%) of our respondents currently perform ECMO as part of their clinical practice. Of those, 83 (91.2%) work at institutions offering both venoatrial (VA) and venovenous (VV) ECMO. Seven (7.7%) and 1 (1%) work at a VA-only and VV-only institution, respectively.

With regard to ECMO management of CDH patients, 73 respondents (68.9%) primarily perform VA ECMO (currently, or previously if there institution does not offer ECMO) and 33 (31.1%) perform VV. Eighty out of 112 respondents (71.4%) have performed both methods of ECMO for CDH, whereas 32 (28.6%) reported never having performed the alternative mode.

When specifically asked which mode of ECMO they believe to be optimal for the management of CDH patients, 46.4% answered VA compared with 53.6% for VV. Those who chose VA ECMO as the more optimal method were 3.3 times more likely to have never performed the alternative method compared with VV proponents, with 35% of VA proponents never having attempted VV and 15% of VV proponents never having performed VA ECMO (OR with 95% CI: 3.3 [1.3, 8.14]).

Primary rationale for VA ECMO preference included cardiac support (57.4%; n = 31), easier management (20.4%; n = 11), and avoidance of high rate of conversion from VV to VA (18.5%; n = 10). The selected rationale for VV ECMO included avoidance of carotid artery ligation (47.5%; n = 28), more physiological support (35.6%; n = 21), avoidance of systemic circulation embolus (8.5%; n = 5), and ease of performance (8.5%; n = 5).

The estimated rate of VV to VA ECMO conversion varied depending on the surgeon's preferred ECMO method. Twenty-one percent of VV proponents reported the rate of conversion to be <5% and only 16% believed it to be >30% compared with 0% and 42%, respectively, in the VA ECMO group (Table 3).

Table 3.

Venovenous to Venoatrial Conversion Rate Estimates Based on Surgeon's Preferred Extracorporeal Membrane Oxygenation Method

Conversion rate (%)	VA proponents, n (%)	VV proponents, n (%)
0–5	0	12 (20.7)
6–15	12 (24)	17 (29.3)
16–30	18 (36)	20 (34.5)
31–50	15 (30)	6 (10.3)
>50	5 (10)	3 (5.2)

VA, venoatrial; VV, venovenous.

Timing of CDH repair

The preferred timing of the respondents for CDH repair in relation to ECMO is reported in Table 4.

Table 4.

Responses for Preferred Timing for Congenital Diaphragmatic Hernia Repair in Relation to Extracorporeal Membrane Oxygenation

Preferred timing of CDH repair in relation to ECMO	Responses, n (%)
Before ECMO cannulation	3 (2.7)
Early during ECMO (within 3 day of cannulation)	25 (22.3)
Once the patient is stabilized on ECMO	17 (15.2)
Late during ECMO (within 3 days of anticipated decannulation)	16 (14.3)
After ECMO decannulation	51 (45.5)

CDH, congenital diaphragmatic hernia; ECMO, extracorporeal membrane oxygenation.

The respondents were then asked whether they had ever performed a CDH repair at an alternative time (a CDH repair at a time in relation to ECMO that is different from their preferred timing). Although the majority of respondents had performed the repair at an alternative time with late ECMO being the most common (53.6%), 21 out of 112 (18.8%) reported having never attempted an alternative timing for repair.

The “last ditch” diaphragmatic repair

When asked whether they would offer CDH repair in a patient failing to improve on ECMO, 71 of 111 respondents (64%) would offer repair at 1 week (15.3%), 2 weeks (28.5%), and 3 weeks (19.8%) of ECMO duration. In contrast, 40 (36%) reported that CDH repair in this setting would be futile. Thirty of 111 respondents (27%) report having witnessed successful decannulation after a CDH repair performed on a patient who had been unable to be weaned after 2 weeks on ECMO.

Exit-to-ECMO

Of 112 respondents, 103 (92%) do not perform exit-to-ECMO in CDH patients. Of the remaining 18 (8%) who do, indications included significant pulmonary hypoplasia/percentage predicted lung volume (PPLV) <15% (n = 14), lung-to-head ratio <1.4 (n = 2), and significant cardiac disease (n = 2).

Discussion

The successful delivery of ECMO support to critically ill CDH patients relies heavily on the multidisciplinary cooperation of a multitude of medical specialties. The surgeon, intensivist, ECMO technician, nursing, and blood bank must all coordinate in perfect synchrony to deliver optimal care. For this very reason, intrainstitutional variation and alteration of ECMO practice are rare to avoid, upsetting the delicate balance and seamless delivery of care to these tenuous patients. This results in conformity of practice, restricting the natural cross-pollination of ideas found in medicine. Unlike in other areas of pediatric surgery, it is not uncommon for a pediatric surgeon to have never witnessed, much less attempt, alternative CDH ECMO practices. This makes one's approach to these patients less a function of comparative evaluation, studies or experience but rather a function of how one was trained. Furthermore, even for the surgeon willing to attempt an alternative approach, the rigid institutional framework and steep learning curves for the other cooperating practitioners would serve as a significant obstacle to implementation. Nowhere is this more apparent than in the very basic question of the optimal mode of ECMO for CDH patients.

Although 73 respondents (68%) primarily perform VA ECMO for cases of CDH, only 45 out of the 73 (62%) believe it to be the optimal approach. This discrepancy was not accounted for by the VA-only ECMO institutions. Furthermore, of the 45 VA performers who believe VA to be optimal, 40% of those had never attempted VV in this setting. This discrepancy between principle and practice is particular accentuated in North America where 56 of 80 respondents (70%) perform VA ECMO for CDH but 25 of those 56 (45%) believe VV to be the optimal method.

Similarly, although the reported rate of VV to VA conversion in the literature is between 0% and 18%,^2–5 VA proponents were more likely to overestimate this rate with 40% believing conversion rate to be >30%. Conversely, only 15% of VV enthusiasts believed the rate to be >30%, with 49% reporting the estimated conversion rate to be <15%.

Timing for CDH repair appears to be another area of controversy. Post-ECMO, CDH repair is presently the most common approach (48%); however, 54% have performed late during ECMO repair at some point in their career, showing a potential shift in trend. Although most respondents had attempted repair at least at one of the alternative times, 18.8% had never attempted repair at an alternative time from which they were trained. This again highlights the uniquely limited exposure to diversity in ECMO management and likely contributes to the difficulty of surgeons to perform adequate studies to evaluate the optimal timing for a CDH repair.

The management of a patient failing to improve on ECMO is one that perpetually plagues the medical community from both a clinical and ethical standpoint. Whereas many believe that repair of the diaphragmatic defect does not improve pulmonary hypertension or enable earlier ECMO decannulation as evidenced by the popularity of the post-ECMO repair approach, our survey demonstrates that, in practice, a majority of pediatric surgeons will attempt the repair in the hopes that it will improve the patients clinical status and enable them to be weaned off ECMO. The successful experience of more than a quarter of the respondents with the “last ditch” CDH repair offers some optimism and guidance in the approach to, what many refer to as, the irrecoverable patient.

Although exit-to-ECMO was initially described for patients with tracheal occlusions and severe CDH,⁶ it is now being used for various anomalies, including lung and neck masses.^7,8 It has also recently been proposed for use in patients with severe CDH with PPLV of <15% on fetal MRI; however, two studies from Children's Hospital of Boston have not shown any significant survival benefit of exit-to-ECMO compared with the more conventional management of CDH.^9,10 Our survey demonstrates that the vast majority of ECMO centers do not participate in this practice.

Although we intended to look for potential practice trends around the world, most of the respondents were from North America, limiting our analysis. The low number of responses, particularly from Africa, South America, and Asia, makes analysis of practice trends from these regions challenging.

ECMO in CDH patients remains an area of controversy and with little consensus. Although low absolute numbers and rigid institutional conformity do not allow for comparative evaluation of different ECMO practices, this survey aims to glean from the clinical experiences of >100 pediatric surgeons in an effort to educate the reader on current practice trends in this complex but critical area.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Khan

, Lally

. The role of extracorporeal membrane oxygenation in the management of infants with congenital diaphragmatic hernia. Semin Perinatol, 2005; 29:118–122.

Heiss

, Clark

. Prediction of mortality in neonates with congenital diaphragmatic hernia treated with extracorporeal membrane oxygenation. Crit Care Med, 1995; 23:1915–1919.

Guner

, Khemani

, Qureshi

, et al. Outcome analysis of neonates with congenital diaphragmatic hernia treated with venovenous vs venoarterial extracorporeal membrane oxygenation. J Pediatr Surg, 2009; 44:1691–1701.

Dimmitt

, Moss

, Rhine

, Benitz

, Henry

MCW

, Vanmeurs

. Venoarterial versus venovenous extracorporeal membrane oxygenation in congenital diaphragmatic hernia: The extracorporeal life support organization registry, 1990–1999. J Pediatr Surg, 2001; 36:1199–1204.

Kugelman

, Gangitano

, Pincros

, Tantivit

, Taschuk

, Durand

. Venovenous versus venoarterial extracorporeal membrane oxygenation in congenital diaphragmatic hernia. J Pediatr Surg, 2003; 38:1131–1136.

Bouchard

, Johnson

, Flake

, et al. The EXIT procedure: Experience and outcome in 31 cases. J Pediatr Surg, 2002; 37:418–426.

Hedrick

. Ex utero intrapartum therapy. Semin Pediatr Surg, 2003; 12:190–195.

Kunisaki

, Barnewolt

, Estroff

, et al. Ex utero intrapartum treatment with extracorporeal membrane oxygenation for severe congenital diaphragmatic hernia. J Pediatr Surg, 2007; 42:98–106.

Shieh

, Wilson

, Sheils

, et al. Does the ex utero intrapartum treatment to extracorporeal membrane oxygenation procedure change morbidity outcomes for high-risk congenital diaphragmatic hernia survivors?. J Pediatr Surg, 2017; 52:22–25.

10.

Stoffan

, Wilson

, Jennings

, Wilkins-Haug

, Buchmiller

. Does the ex utero intrapartum treatment to extracorporeal membrane oxygenation procedure change outcomes for high-risk patients with congenital diaphragmatic hernia?. J Pediatr Surg, 2012; 47:1053–1057.