Section 9: Home mechanical ventilation in children with kyphoscoliosis

This article is related to:

Pediatric home mechanical ventilation: A Canadian Thoracic Society clinical practice guideline executive summary

Introduction

Kyphoscoliosis is a well-recognized cause of respiratory failure in adults.¹ In pediatrics, there is a growing body of literature on respiratory failure secondary to early onset scoliosis associated with other bony abnormalities of the thorax. Most children with scoliosis who require long-term assisted ventilation have neuromuscular scoliosis, in which case, although scoliosis may contribute to respiratory compromise, the requirement for assisted ventilation is usually mandated by the underlying condition.^2,3 Rotation of the thoracic vertebrae impairs the upward displacement of the ribs during inspiration and places the respiratory muscles at a mechanical disadvantage. In addition, there is displacement and crowding of the organs of the thoracic cavity on the concave side with resulting decreases in both chest wall compliance and lung compliance.⁴

Surgical spinal stabilization is an integral component of the treatment of scoliosis, and is regularly performed prophylactically to preserve respiratory function, prevent the future development of pulmonary hypertension, improve quality of life by facilitating seating and for cosmetic reasons (to improve symmetry on standing).^5,6

Historically, spinal surgery was denied to children at risk of not being able to be weaned from supportive ventilation postoperatively. For example, based on the Duchenne’s muscular dystrophy (DMD) literature,^7–15 many patients with scoliosis were deemed ineligible for surgery if the forced vital capacity (FVC) was less than 30–40% of predicted values. Some surgeons in the past advocated mandatory preoperative tracheostomy in patients with severe restrictive disease.^7–15 With the introduction of noninvasive ventilation (NIV), both for immediate postoperative as well as long-term ventilatory support, in addition to improved anesthetic and operative techniques, such limitations have been reexamined.

It is therefore important that a guideline not only address the question of providing home mechanical ventilation (HMV) to pediatric patients with respiratory failure and scoliosis, but also the eligibility for corrective scoliosis surgery, the consequences of surgery on pulmonary function and discussion of techniques to reduce morbidity following surgery.

Literature review: Methodology

Searches were conducted looking for publications on (1) scoliosis, kyphoscoliosis; (2) pulmonary function tests; and (3) ventilation. We aimed at identifying all studies published in English and French. Cochrane and MEDLINE databases (1966–August 24, 2015) were searched. As well, reference lists from identified publications were hand searched in order to add any missed studies. We also searched the web sites of large associations of physicians and health professionals in the fields of respiratory medicine, intensive care, nursing and respiratory therapy for reviews, consensus statements and clinical practice guidelines. We obtained the full publication on all relevant studies identified.

Results

We retrieved 58 English-language publications that were relevant to our inclusion criteria and dealt with home mechanical ventilation in children,. Editorials and opinions, general reviews, publications with clinical practice guidelines and publications on the type of equipment used for airway clearance were eliminated, as were publications dealing solely with adult populations. We were left with 32 publications dealing specifically with cohorts including children with scoliosis and/or kyphoscoliosis and/or on HMV.

Discussion

Provision of HMV to patients with respiratory failure and scoliosis is associated with long-term benefits

There is a very limited pediatric literature on primary outcomes of long-term ventilation in children with isolated or neuromuscular disease-associated scoliosis, and further research is needed (see Table 1). Leger published a large study of 276 patients with scoliosis managed with NIV. However, only the DMD subgroup of 16 patients included children.

Predictors of prolonged mechanical ventilation post-surgical scoliosis repair

We found 11 retrospective studies reporting on 592 pediatric patients undergoing spinal fusion surgery, of whom 134/558 (24%) required prolonged mechanical ventilation, defined as ≥36 hours (see Table 2).^16–26 It was unclear how many of the patients in the paper by Mills et al. remained extubated at ≥36 hours.¹⁶ Twenty-three (2.5%) of the patients received HMV preoperatively.

All of the studies included children with preoperative forced vital capacity (FVC) percent predicted less than 30–40%. Three studies found that lower preoperative FVC% predicted significantly predicted the need for prolonged mechanical ventilation.¹⁷ However, the precise values of FVC that predict the need for postoperative ventilation remain unclear; all 3 of these studies incorporated different FVC cutoffs. Jenkins et al. found that prolonged mechanical ventilation only occurred in patients with FVC less than 40% predicted.^18,19,21 Udink et al. reported the mean (range) FVC% predicted in those with prolonged mechanical ventilation was 34 (21–25)% as compared to 62 (30–114)% in those not needing prolonged mechanical ventilation.¹⁸ Finally, Yuan et al. found that FVC <60% predicted was the most significant predictor of prolonged mechanical ventilation.¹⁹ In contrast, in 2 other studies, preoperative FVC% predicted was not a significant predictor of prolonged mechanical ventilation. Two studies found that a lower forced expiratory volume in 1 second (FEV₁) significantly predicted the need for prolonged mechanical ventilation, but 1 reported no significant association.^18,19,24 In those patients unable to perform pulmonary function tests, 1 retrospective study reported on the use of infant pulmonary function testing and PSG to predict the need for mechanical ventilation post-spinal fusion, but found no relationship between the two.²⁵

Other predictors of prolonged mechanical ventilation reported to be significant by Yuan et al. included older age, male gender and the presence of scoliosis due to neuromuscular disease.¹⁹ These findings were not reproduced by other studies. Zhang et al. found surgical approach to be the only predictor of prolonged mechanical ventilation in their study,²⁴ while Gurajala et al. found longer vertebral fusions and hypothermia to be associated with prolonged mechanical ventilation.²⁶

While there is evidence that the degree of impairment of pulmonary function correlates with the need for prolonged mechanical ventilation post-scoliosis repair, there does not appear to be a consensus on the defined cutoff level that best predicts the need for protracted ventilatory support.

Pulmonary function post-surgical scoliosis repair

Pulmonary function following scoliosis surgery has been studied comprehensively in patients with idiopathic scoliosis but not in patients with neuromuscular scoliosis (see Table 3).^27–39 Following scoliosis surgery, pulmonary function decreases dramatically in the immediate postoperative period,³¹ but improves during the weeks following surgery, with resolution of postoperative secretion retention and atelectasis. At 3 months following surgery, many patients will have returned to baseline values, and most will reach their preoperative baseline by 2 years postoperatively.^27,29 However, many patients, particularly those whose surgery included thoracotomy and those patients with more advanced restrictive lung disease, may have long-term decreased pulmonary function.³⁴

New surgical techniques incorporating thoracoscopic anterior approaches may reduce morbidity and have fewer detrimental effects on pulmonary function (see Table 4).^32–34 More recent studies have demonstrated an improvement in lung function at 2- and 3-year follow up.^38,39 Huitema et. al. demonstrated a significant improvement in FEV₁ in 20 adolescents with scoliosis at 2-year follow up, as compared to baseline.³⁸ Demura et. al. demonstrated similar results in a prospective cohort of 154 adolescents.³⁹ However, the increases in absolute lung function values that are reported are likely due to growth rather than true improvement. There was no change in FEV₁% predicted. This highlights the advantages of less-invasive surgical repairs in both the peri- and postoperative periods, as well as in long-term follow up.

For patients with congenital thoracic scoliosis, use of the vertical expandable prosthetic titanium rib device (VEPTR) has been shown to augment chest cavity size, but the effect on pulmonary function is not clear.

Scoliosis patients who have not previously been on assisted ventilation may benefit from NIV following extubation.^27,31 Since the introduction of postoperative NIV, failure to wean from invasive ventilation is rare. Most patients not receiving assisted ventilation prior to surgical correction will be able to discontinue NIV prior to hospital discharge, although a small proportion may need to continue on assisted ventilation at home. Preoperative NIV and assisted cough techniques, in addition to routine extubation to NIV post-procedure, will result in a reduced incidence of failure to wean from invasive ventilation, and may shorten hospital stays.^27–32

Conclusion

In summary, the provision of HMV to patients with respiratory failure and scoliosis is associated with long-term benefits. From the available literature, an FVC value is not clearly predictive of the need for prolonged ventilation postoperatively, nor can a meaningful estimate of the percentage of patients requiring mechanical ventilation postoperatively following scoliosis repair be determined. Mandatory pre-operative tracheostomy is clearly no longer essential. This does not preclude a frank discussion with the patient and family regarding the risks of spinal surgery including the possibility of prolonged ventilation and tracheostomy in at-risk patients. It is also apparent that, despite the effects of stabilization on the spine as a result of spinal fusions, some patients may need to be maintained on HMV.

Research questions

1. What is the definitive FVC% predicted or FEV₁% predicted that predict the need for prolonged mechanical ventilation in the postoperative period?

2. Does the addition of regular lung volume recruitment and airway clearance techniques as well as other secretion management interventions affect the surgical outcomes of patients with kyphoscoliosis?

3. What short- and long-term effects will the use of the vertical expandable prosthetic titanium rib device (VEPTR), as well as all growth-sparing techniques, have on pulmonary function in patients with complex thoracic scoliosis?

Recommendations for long-term mechanical ventilation at home in children with kyphoscoliosis

1. Long-term mechanical ventilation is recommended for all patients with kyphoscoliosis who develop chronic respiratory failure. (Grade 1C)

2. Preoperative pulmonary function inversely correlates with the need for prolonged mechanical ventilation post spinal fusion therapy. However, there is no consensus on a definitive FVC% predicted or FEV₁% predicted that predicts the need for prolonged postoperative mechanical ventilation. (Grade 2C).

3. Children should not be refused corrective scoliosis surgery on the basis of pulmonary function testing alone. (Grade 2C)

4. A tracheostomy is not mandatory for spinal fusion surgery. (Grade 2C)

5. The initiation of NIV should be considered in the preoperative period. (Grade 2C)

6. There should be a low threshold for the initiation of NIV after extubation following spinal corrective surgery. (Grade 1C)

7. Airway clearance techniques (breathstacking with and without assisted cough or mechanical in-exsufflation) should be provided in the postoperative period following complex spinal surgery in patients with neuromuscular scoliosis. (Grade 1C)

Table 1. Provision of HMV to patients with respiratory failure and scoliosis is associated with long-term benefits: Summary of evidence.

Author (year)	Study type	Number of patients	Age (years)	Scoliosis type	Primary outcomes of long-term ventilation in scoliosis
					survival	ABGs	PFTs	Hospital days	QoL
Leger (1994)^40	retrospective	276 NIV	mean (SD) 57 (13)	idiopathic (105/276)	N/A	PaO2 and PCO2 improved and improvements maintained for ≥2 years	No significant change in TLC, FVC and FEV1/FVC	Significantly lower (P < 0.0001) Most pronounced in kyphoscoliosis group (34 ± 31 days to 6 ± 6 days)	Sleep improved (62%) Increased ability to carry out daily activities (70%) Nasal and mouth dryness (50%) GI distension from swallowing air (50%)
Fauroux (2003)^2	Retrospective pediatric questionnaireretor	102 (8 scoliosis)	7% < 3 years 6% 3–5 years 29% 6–11 years 58% 12–18 years	not specified	N/A	Only in 76% of patients	N/A	N/A	NA
Appierto (2002)^3	retrospective	46 (4 congenital scoliosis)	0–18 years of age	congenital	6/46 (13%) died	N/A	N/A	N/A	N/A

ABG: arterial blood gas; CO₂: carbon dioxide; FVC: forced vital capacity; NIV: noninvasive positive pressure ventilation; PFTs: pulmonary function tests; QoL: quality of life.

Table 2. Predictors of prolonged mechanical ventilation post-surgical scoliosis repair: Summary of evidence.

					Predictors of the need for prolonged mechanical ventilation
Author (year)	Study type	No. of patients	Age (years)	Scoliosis type	Age	Sex	BMI	Scoliosis etiology	Surgical approach	Pre-op FEV1	Pre-op FVC	# Needing prolonged mechanical ventilation
Jenkins (1982)^17	Retrospective	50	Mean (range) 13 (7–21)	mixed	N/A	N/A	N/A	N/A	N/A	N/A	FVC all <40% predicted	Ventilation >36 hours: 4 (8%)
Udink (2008)^18	Retrospective	46	Median (IQR) 3–17(13.5)	Neuro-muscular	N/A	N/A	N/A	N/A	N/A	Lower 33% (21–57) versus 67% (22–114)	Lower 34% (21–55) vs 62% (30–114)	Ventilation ≥72 hours: 7 (15%)
Yuan (2004)^25	retrospective	125	Mean (SD) 13.7 (3)	mixed	Older OR =7 P = .001	male	N/A	neuro muscular	N/A	Lower FEV1% predicted <40% most specific predictor of MV	Lower FVC% predicted <60% most sensitive predictor of MV	Ventilation ≥72 hours: 40 (32%) (preoperatively 6 were on NIV
Wazeka (2004)^20	retrospective	21	Mean (range) 11.5 (2–21)	Mixed	N/A	N/A	N/A	N/A	N/A	N/A	N/A	HMV: 6 (29%) (Preoperatively 4 were on NIV
Payo (2009)^21	retrospective	24	Mean (range) 13 (9–19)	Mixed	N/A	N/A	N/A	N/A	N/A	N/A	N/A	17 needed prolonged MV (71%) (10 were on preoperative home NIV)
Marsh (2003)^22	retrospective	30	Mean (range) 14 years (8 months–19 years)	Neuromuscular (DMD)	N/A	N/A	N/A	N/A	N/A	N/A	N/A	Ventilation >72 hours: 25 (20%)
Rawlins (1996)^23	retrospective	32	Mean (range) 13 (7–17)	Mixed	N/A	N/A	N/A	N/A	N/A	N/A	N/A	HMV via tracheostomy: 4 (12.5%) (1 patient had preoperative tracheotomy)
Zhang (2005)^24	retrospective	298	Mean (range) 16.4 (6–62)	Mixed	N/A	N/A	N/A	N/A	see comment	N/A	N/A	Postoperative ventilation: 6 (2%) Transthoracic approach increased complications
Yuan (2005)^31	retrospective	128	Mean (SD) 10.8 (4.9)	Mixed	N/A	N/A	N/A	N/A	N/A	N/A	N/A	Ventilation >72 hours: 25 (20%)
Gurajala (2013)^26	retrospective	102	Mean (SD) 14.31 (3.78)	Mixed	N/A	N/A	N/A	N/A	Posterior fusion: 93% Anterior: 2.9% Combined: 3.9%	N/A	N/A	Longer fusions of vertebrae (> 8) (OR 1.29) and Hypothermia (OR .096) were significantly associated with post-op. mechanical ventilation
Mills (2013)^16	retrospective	34	Mean(range) 11.7(6.6–16.6)	Neuromuscular	N/A	N/A	N/A	N/A	N/A	N/A	15 had preop FVC <40%	All extubated by POD #3

BMI: body mass index; FEV₁: forced expiratory volume in one second; MV: mechanical ventilation.

Table 3. Pulmonary function post-surgical scoliosis repair: Summary of evidence.

				Pulmonary function post surgical scoliosis repair
						Postoperative pulmonary function
Author (year)	Study type	Number of patients	Age (years)	Scoliosis	Surgical approach	Early	3 months	1 year	2.5 years
Lenke (1995)^27	Prospective	26 pediatric 7 adult	Mean 13.9	idiopathic	posterior fusion with thoracoplasty	NA	−16%	−6%	Back to baseline for pediatrics and -23% adult
Graham (2000)^28	prospective	51	Mean (range) 15.0 (0.92–20.4)	idiopathic	open thoracotomy	NA	−19%	−8%	−5%
Vedantam (2000)^29	prospective	98	Mean (range) 14.6 (10–20)	idiopathic	no thoracotomy (n = 47) thoracoplasty (n = 33) thoracotomy (n = 7) combined (n = 11)	NA	0 (?) −13% −20% −8%	−13% −10% −10% −9%	−6% −5% −5% −3%
Kim (2005)^30	prospective	118	Mean (SD) 14.4 (1.9)	idiopathic	thoracoplasty	NA	NA	NA	−7%
Yuan (2005)^31	prospective	24	Mean (SD) 12.7 (2.7)	Idiopathic 11 Congenital 2 Neuromuscular 9	Thoracotomy (n = 9)	up to 60% decline in FVC	NA	NA	NA
Lonner (2007)^32	prospective	51	Mean 14.5	idiopathic	thoracoscopic (n = 28) posterior fusion (n = 23)	NA	NA	NA	Baseline in both groups
Newton (2007)^33	prospective	254	Mean (SD) females: 14 (2) males: 15 (2)	idiopathic	no thoracotomy (n = 53) thoracoscopy (n = 40) thoracotomy (n = 158)	NA	NA	NA	+1% −1% −15%
Kishan (2007)^34	prospective	107	Range 12–18	idiopathic	thorascopy (n = 36) open thoracotomy (n = 61)	NA	NA	NA	−2 % −15 %
Karol (2008)^35	prospective	28	Mean 3.3	Congenital 20 Idiopathic 3 Other 5	thoracic fusion	NA	NA	NA	2–11 years postop FVC, < 50%
Campbell (2003)^36	prospective	21	Mean 3.3	congenital	VEPTR	NA	NA	NA	NA
Gadepalli (2011)^37	prospective	26	Mean (SD) 7.6 (3.4)	Congenital neuromuscular	VEPTR	NA	NA	same	NA
Huitema (2013)^38	retrospective	20	Mean (SD) 16 (3,1)	Adolescent thoracic idiopathic scoliosis	Minimal open anterior spinal fusion	Mean FEV1: 2.81	N/A	N./A	Mean FEV1: 3.14
Demura (2013)^39	prospective	154	Mean (SD) 14.6 (2.2)	Adolescent idiopathic scoliosis	Posterior fusion	FVC (L): 2.90 ± 0.76 FEV1 (L): 2.45 ± 0.63	N/A	N/A	FVC (L): 3.17 ± 0.78 FEV1 (L): 2.71 ± 0.68
Mills (2013)^16	retrospective	34	Mean (range) 11.7(6.6–16.6)	Neuromuscular scoliosis	Posterior fusion	FVC % first post-op visit compared to pre-op 34 (5–52) vs 39 (5–55)	N/A	N/A	N/A

Table 4. Reduction of morbidity post-surgical scoliosis repair: Summary of evidence.

Author (year)	Study type	Number of patients	Age (years)	Scoliosis type	Primary outcomes of long-term ventilation in scoliosis					Comments
					survival	ABGs	PFTs	Hospital days	QoL
Yuan (2005)^31	retrospective	125	Mean (SD) 12.7 (2.7)	idiopathic (27) congenital (13) neuromuscular (57	100%	N/A	FVC <60% predicts >3 PICU days	N/A	N/A	Predicting prolonged postop PICU care
Bach (2005)^41	prospective	5	Range 9–13 year	neuromuscular	100%	N/A	FVC <40%	7–14 days	N/A	scoliosis surgery in patients with high pulmonary risk – supportive techniques
Hod-Feins (2007)^42	retrospective	126	Mean (range) 14.3 (9.5–22)	idiopathic (95) neuromuscular (31)	100%	N/A	N/A	Mean 3.8 PICU days	N/A	delayed extubation (17/126)
Bach (2010)^43	prospective	157	Mean (SD) 37 (21)	neuromuscular	100% 17 late deaths	N/A	N/A	N//A	N/A	5 new tracheostomies required
Takaso (2010)^44	prospective	10	Mean (range) 13 (11.2–17.3)	neuromuscular	100%	N/A	Pre, postop 6 weeks, every 6 months	N/A	N/A	Preoperative preparation – inspiratory muscle training 2 year follow up
Vianello (2011)^45	letter to editor	N/A	Mean (SD) Experimental: 12.19 (23) Control: 20 (35)	neuromuscular	N/A	N/A	N/A	N/A	N/A	Labour intensity – training of family members