Introduction
Coarctation of the aorta is a relatively common congenital defect, affecting 1 in 2500 live births. Corrective surgery was first performed in 1944 and a variety of techniques have been employed to relieve the stenosis, with now excellent post-operative survival.Reference Adams, Davidson and Swallow1–Reference Cohen, Fuster, Steele, Driscoll and McGoon4 The subclavian flap repair, in which the subclavian artery is ligated distally and the proximal end reattached over the coarctation to relieve the obstruction,Reference Waldhausen, Petry and Kurlander5 has been used for decades. Despite this, there have been concerns about the possible long-term consequences of this on the left arm.Reference Pandey, Jackson, Ajab, Gladman and Pozzi2 Reported effects on vascular supply, strength and function have been variable and limited by subjective self-reporting,Reference Adams, Davidson and Swallow1, Reference Barreiro, Ellison and Williams6 child or relatively short-term follow up.Reference Shenberger, Prophet, Waldhausen, Davidson and Sinoway7–Reference van Son, van Asten and van Lier9 To date, research on long-term structural and functional sequelae of subclavian flap repair in an adult population has been scant and limited to non-standardised data collection processes.
We aimed to analyse upper limb structural and functional parameters in adults who had surgery in childhood for coarctation of the aorta using either subclavian flap repair or end to end anastomosis. In this latter group, vascular supply to the left arm is unaffected. We examined between-limb differences, in right-handed patients, comparing the consequences of the two different types of coarctation repair procedure.
Methods
Study population
This single centre prospective study was performed at a quaternary level adult congenital heart disease centre in Sydney, Australia. Through our adult congenital heart disease database, patients >16 years of age with coarctation of the aorta repaired with either subclavian flap repair or end to end anastomosis were identified. Those with concomitant complex CHD were excluded and only right-handed subjects were included. Patients were consecutively contacted in alphabetical order and recruited into the study until a total of 18 patients were consented and enrolled (9 subclavian flap, 9 end to end anastomosis). Patients with re-intervention on the coarctation after initial repair, or significant residual obstruction (defined as blood pressure arm versus leg difference of >20 mm Hg) were excluded. The study was completed under the Sydney Local Health District, Ethics Research and Governance, Protocol Number X17-0012.
Study design
A prospective observational two-group design was used to compare structural (muscle mass, bone length and blood flow) and functional attributes of these adults. All patients underwent MRI, ultrasound imaging of upper extremity blood flow and functional assessments, as outlined below.
Outcomes: We prospectively defined one primary structural and one primary functional endpoint. The primary structural outcome was the difference in forearm muscle mass between right and left arms, comparing differences between the subclavian flap and end to end anastomosis groups. It was hypothesised that this right–left difference would be significantly greater in patients after subclavian repair compared to end to end anastomosis. The primary functional endpoint was the difference in grip strength between right and left arms between groups. Secondary outcomes included differences in bone length, blood flow, grip strength and functional metrics, between the right and left arms comparing the two groups.
Power and recruitment: Previous studies have shown a 4% average difference in muscle thickness associated with hand dominance.Reference Abe and Loenneke10 Therefore, power calculations were made assuming a 5% difference in forearm muscle thickness (right minus left) in end to end anastomosis and a 20% difference (right minus left) in the subclavian flap group. A 10% SD for right–left difference was assumed in each group. To achieve 80% power at the 2p < 0.05 significance level, seven subjects per group were required. Allowing for drop-outs, technically inadequate scans and lack of precision around the assumptions above, we aimed to enrol nine subjects per group.
General assessment
Age, height, weight and hand dominance were ascertained. Past medical history was taken assessing for the presence of hypertension, diabetes, smoking history, self-reported substance abuse or any other medical condition that could impact arm functioning.
MRI protocol for bone length and muscle mass assessment
MRI scans were taken using a 1.5 T Phillips imaging unit at specialist MRI (sMRI). The patient lay supine with arms extended at their side, 2 × 4 Channel Flex Coils (1 Large 1 Medium Size) and the in-built bore body coil were used for the upper arm. An 8 Channel SENSE Body Coil was utilised for the lower arm. Localiser images were obtained. For bone length measurement and anatomical land marking, T1-weighted coronal images (FOV 400, slice thickness 3 mm, 1 mm Gap, TR 430-650, TE 10) of the upper arm and sagittal T1-weighted images of the forearm were completed (FOV 350, slice thickness 3 mm, 1 mm gap, TR 430-650, TE 10). For muscle measurement of the upper arm; cross-sectional T1-weighted images using a spin echo sequence were obtained (FOV 200, slice thickness 10 mm, no gap, TR 400-700, TE 10) from the mid-point of the elbow joint to 15 cm proximally. For the lower arm, cross-sectional images T1-Weighted images (FOV 150, slice thickness 10 mm, no gap, TR 400-700, TE 10) were obtained from the metacarpal phalangeal joints to above the olecranon process. Data was then analysed using Osirix MD v9.5.1, Pixmeo Sarl, Bernex, Switzerland. All MRI analyses were completed by a trained cardiac magnetic resonance imaging cardiologist, blinded to patient, hand dominance, demographics and type of coarctation repair (MD). Inter-rater reliability was assessed by RP completing two studies blinded with a difference <5%.
Bone length measurements: Bone length measurements (humeral, radial and ulnar) were taken using MRI in the equivalent positioning to gross specimen measurements as per established practice – Buikstra and Ubelaker.Reference Skinner, Buikstra and Ubelaker11
Muscle mass measurements: On each slice, muscle tissue boundaries were traced, with bone and fat excluded. In the upper arm, slices were included for muscle mass measurement from 15 cm proximal to the mid-point of the elbow joint to, but not including the first slice where the head of radius was visualised. For the forearm, the first slice including head of the radius to the radiocarpal joint was included for muscle mass measurement. Muscle volume was calculated by the summation of areas of all the slices automatically taking into account the slice thickness. Maximal cross-sectional area (MCSA) was defined as the maximal slice area measured, excluding bone and adipose tissue.
Ultrasound blood flow protocol
Vascular ultrasound was used assess blood flow velocity and endothelial function through the left and right arms using a modified flow mediated dilatation (FMD) protocol as previously described.Reference Corretti, Anderson and Benjamin12 Patients were asked to refrain from caffeine, 24 hours prior to the procedure. The examination was performed in a temperature-controlled room (20–25°C) with subjects resting in a supine position. Blood pressure measurements were taken a minimum 10 minutes prior to commencement of tests. Vessel flow was assessed using ultrasound measurement of the brachial artery diameter and changes in brachial artery flow (7–12 MHz linear array, Logic 7, GE). In subclavian flap repairs in the left arm, the largest vessel at or proximal to the antecubital fossa was used. A 9 cm pneumatic cuff is placed around the forearm immediately below the elbow and a 4 cm length of the brachial artery was imaged in a longitudinal section above the antecubital fossa. Arterial diameters and flow velocities were recorded before and after cuff inflation/deflation.
Physical capability and functional performance assessment
Functional performance was indicated by using standardised tests of strength, manual dexterity, bimanual coordination and unilateral gross manual dexterity. All testing was conducted by registered occupational therapists (AC & JB) who trained in the protocol to ensure consistent administration and scoring. Inter-rater agreement was assessed across all measures with 5 of the 18 participants having data collected simultaneously by both raters, independently scored and results compared.
Physical capability – strength: Handgrip strength was assessed using the Jamar™ hydraulic hand dynamometer with a standardised protocol as per guidelinesReference Roberts, Denison and Martin13 including self-selected handgrip, standing position with the same leg-stance, forearm and elbow position across participants and within trials. The same dynamometer was used across all participants. The mean score of three trials was used as the final score for each limb. Inter-rater agreement was 100%.
Fine motor manual dexterity: The Jamar™ Nine Hole Peg Test (NHPT)Reference Kellor, Frost, Silberberg, Iversen and Cummings14 model of this standardised timed observed test of finger dexterity was used.Reference Oxford Grice, Vogel and Le15 Participants completed the test three times using alternating limbs. The mean time was used as the final score for each limb. The time was read off the digital stopwatch which showed minutes and seconds to the third decimal place; one decimal place was used. Test time commenced with the verbal instruction to start and finish time was taken when the last peg hit the holding-dish. Inter-rater agreement using full seconds only was 100%.
Unilateral gross-motor dexterity: The standard Box and Blocks test (BBT) is an observed timed performance test that measures unilateral gross manual dexterityReference Mathiowetz, Volland, Kashman and Weber16 and is one of the most commonly used assessments of upper limb function across a wide variety of clinical populations.Reference Kontson, Marcus, Myklebust and Civillico17 A commercially available wooden BBT was used (Sammons Preston Model). Participants completed the test three times using alternate limbs. The number of blocks moved correctly in each trial was counted and the mean taken as the final score. Inter-rater agreement was 100%.
Manual dexterity and bi-manual coordination: The Perdue Peg Board Test (PPB)Reference Tiffin and Asher18 is a standardised observed timed performance test measuring fine and gross movements of the fingers, hands and arms as well as fine fingertip dexterity. A commercially available PPB was used (Lafayette Instrument Model 32020A). Between one and three complete PPB trials were completed by each participant; variation in trial number was a result of testing scheduling. The mean of trials completed was used for final scores.
Activity participation: Activity participation was assessed using the Quick-DASHReference Kennedy, Beaton and Smith19 standardised self-report measure developed to describe disability experienced by respondents and monitor symptoms and function. The freely available Quick-DASH disability/symptoms section (11 items) was obtained and used in accordance with conditions of use.
Statistical analysis
Statistical analysis was completed using SPSS version 25.0 (IBM Inc., Armonk, NY, United States of America). Continuous variables were summarised as mean and standard deviation or median with inter-quartile range. Student t-test was used for comparison of continuous variables. Categorical variables were summarised as frequencies and percentages. Comparison of categorical variable groups was performed using chi-squared test or Fischer exact test, where applicable. A two-tailed value of p < 0.05 was considered statistically significant.
Results
All 18 patients (38 ± 12 years, 78% males) were independent in activities of daily living and underwent ultrasound and functional assessment. Seventeen (nine subclavian flap and eight end to end anastomosis) completed MRI assessment (one subject was excluded owing to presence of an implantable cardiac defibrillator). The mean age at coarctation repair was 4.7 ± 5.9 years (range 0.1–18 years); and was similar between groups (subclavian flap 3.9 ± 5.9 years, end to end anastomosis 5.9 ± 6.2 years, p = 0.62). The mean time from initial repair was 32 ± 9 years and there was no significant difference in follow-up periods between groups (33 ± 6 years – subclavian flap, 32 ± 12 years – end to end anastomosis; p = 0.73). At the time of study, no patient had significant morbidity that would impact study assessment. No patient reported symptoms suggestive of ischaemia on rest or exercise.
Overall cohort
Right arm blood pressure, bone length and muscle mass metrics were all significantly greater than the left arm in all subjects (Supplementary Table 1), except for radial bone length and the percentage change in the FMD.
End-to-end anastomoses versus subclavian flap repair (between-group comparison)
Full Comparison of the end to end anastomosis cohort and subclavian flap cohort is shown in Table 1. There were no significant differences in the bone or muscle metrics between the end to end anastomosis and subclavian flap cohorts. Mean diastolic blood pressure of both arms were significantly higher in the end to end anastomosis group in right (84 ± 9.6 mm Hg versus 75 ± 10 mm Hg; p = 0.017) and the left arms (79 ± 12.5 mm Hg versus 66 ± 9.0 mm Hg; p = 0.018).
Table 1. Comparison of parameters in SF versus EEA.
MCSA = maximal cross-sectional area; DASH = disabilities of the arm, shoulder and hand; QDASH = quick DASH.
Within-group differences comparison: end to end anastomoses versus subclavian flap cohorts
Differences observed between limbs, between groups are reported in Table 2. The subclavian flap group had a significantly larger difference in lower arm muscle mass – right greater than left (94.5 ± 42.3 mls versus 37.7 ± 94.5 mls, p = 0.008), greater arm maximal cross-sectional area (right greater than left), (5.9 ± 2.8 cm2 versus 2.9 ± 2.6 cm2, p = 0.038). In the subclavian flap group, right versus left grip strength was also significantly greater than in the end to end anastomosis group (14.7 ± 8.3 versus 5.9 ± 5.3, p = 0.016) between the right and left arms.
Table 2. Right versus left arm differences in structure and function by coarctation repair type.
MCSA = maximal cross-sectional area; PPT = purdue peg test; NHP = nine-hole peg test.
Functional assessment
Functional dexterity test results between repair types are shown in Table 1. There were no other significant differences between the two groups.
Discussion
To our knowledge, this study provides the most complete structural and functional assessment of the upper limbs in adults who have undergone anatomically successful subclavian flap or end to end anastomosis for coarctation repair in childhood. We have found significantly greater right versus left arm muscle mass and grip strength differential in the subclavian flap repair patients, compared to the difference in right versus left arms in end to end anastomosis subjects.
The structural and functional consequences of left subclavian artery ligation in the treatment of congenital heart disease, whether it be Blalock Taussig shunt or subclavian flap repair, have predominantly been reported in paediatric populations.Reference Shenberger, Prophet, Waldhausen, Davidson and Sinoway7, Reference van Son, van Asten and van Lier9, Reference Currarino and Engle20–Reference Todd, Dangerfield, Hamilton and Wilkinson22 Reduced blood flow at rest and during hyperaemia,Reference van Son, van Asten and van Lier9 reduced longitudinal bone length Reference Todd, Dangerfield, Hamilton and Wilkinson22, Reference Lodge, Lamberti and Goodman23 and muscle thickness, as measured by anthropometry, have been reported in subjects who have undergone subclavian flap repair.Reference Currarino and Engle20 Abnormalities in blood flow, arm length and grip strength have also been demonstrated in ipsilateral arm of patients with Blalock Taussig shunts.Reference Skovranek, Goetzova and Samanek21, Reference Harris, Segel and Bishop24 However, these results have not been uniformly demonstrated, with Shenberger et alReference Shenberger, Prophet, Waldhausen, Davidson and Sinoway7 finding no difference in blood flow nor static forearm strength, 9 years after subclavian flap repair.
In adults with repaired coarctation, evidence on the structural and functional sequelae of left subclavian ligation is scarce and limited by subjectivity. Self-reported differences in arm length and muscle development post subclavian flap repair have been shown in someReference Pandey, Jackson, Ajab, Gladman and Pozzi2, Reference Barreiro, Ellison and Williams6 but all studies.Reference Adams, Davidson and Swallow1
We found significant differences in right versus left lower arm muscle mass and grip strength of patients with subclavian flap repair compared to the differences seen in end to end anastomosis, whilst no significant differences were noted in the upper arm, suggesting a preferential effect on the left lower arm of ligation of the subclavian artery. This is in contrast to the findings of Van Son et alReference van Son, van Asten and van Lier9 who reported significant lower blood flow velocity, FMD, upper arm length and circumference in the left arm of subclavian flap repaired children (median age 8 years, IQR 2.5). These discordant findings likely relate to a number of factors. Our cohort’s mean age (38 ± 12 years) is substantially older, and thus further from initial repair, and a reduction in blood flow abnormalities is known to occur with increased time from initial surgery,Reference Skovranek, Goetzova and Samanek21, Reference Todd, Dangerfield, Hamilton and Wilkinson22 presumably with the development of collateral blood supply. Therefore, it is unlikely time of initial repair had a significant impact. Collateral blood supply has also been demonstrated to be influenced by the location of subclavian artery division.Reference van Son, van Asten and van Lier9 We did not assess the specific site of division on all our patients. Further, previous studies have utilised anthropometry for bone length and muscle circumference, where we utilised MRI for more accurate comparison of these parameters.
Our study reported higher diastolic blood pressure in end to end anastomoses patients when compared to the subclavian repair cohort and a non-significantly higher systolic blood pressure. This is in contrast to previous studies reporting higher blood pressures in subclavian flap repair cohorts.Reference Canniffe, Ou, Walsh, Bonnet and Celermajer25 The aetiology of late-onset hypertension in coarctation patients is thought to be multi-factorial including age of repair, type of repair, activation of the rennin–angiotensin system, residual coarctation, aortic arch type and intrinsic abnormalities in the aortic vasculature.Reference Vigneswaran, Sinha, Valverde, Simpson and Charakida26 Our study was not sufficiently powered to detect meaningful differences nor causes of blood pressure differentials between groups.
We performed rigorous standardised functional testing of arm function and found no significant differences between subclavian flap repair and end to end anastomosis subjects when comparing differences between right arm and left arm function. This is in contrast to previous studies reporting subjective differences without any standardised objective testing.Reference Pandey, Jackson, Ajab, Gladman and Pozzi2 It is likely that the lack of functional difference relates to learning and plasticity of motor skills but is possibly due to the insensitivity of tests applied, to pick up small functional differences between the right and left arm. Furthermore, arm function and dexterity are likely related to occupation and training effects, not only to anatomical and physiological structure.
Limitations
The assessment of upper limb muscle volume did not include the whole length humerus and shoulder, however the 15 cm proximal to the elbow joint did include the maximal biceps muscle volume and area. Blood flow was only assessed in one artery in each group and regional blood flow variations may exist and were not assessed. Left-handed subjects were not studied, and so our observations may not be generalisable to this group.
Conclusion
Subclavian Flap repair results in significant differences in muscle mass and grip strength between the right and left arms when compared to patients with end to end anastomosis in adults with repaired coarctation of the aorta.
Supplementary Material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1047951119000386.
Author ORCIDs
Mark R. Dennis 0000-0002-1281-1324
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Disclosures
The authors report no conflict of interest. These authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. The authors report no relationship to industry related to this study.
Ethics Approval
The study protocol was approved by the institutional review board of Royal Prince Alfred Hospital, Sydney Local Health District.
Contributors
All authors have read and approved the manuscript. None of them have declared conflict of interest. Each author has contributed significantly to the work.
Provenance and Peer Review
Not commissioned; externally peer reviewed.
