Differential Findings on Anorectal Manometry in Patients with Parkinson's Disease and Defecatory Dysfunction

Abstract

Introduction

Gastrointestinal dysfunction, particularly constipation, is among the most common non‐motor manifestations in Parkinson's Disease (PD). We aimed to identify high‐resolution anorectal manometry (HR‐ARM) abnormalities in patients with PD using the London Classification.

Methods

We conducted a retrospective review of all PD patients at our institution who underwent HR‐ARM and balloon expulsion test (BET) for evaluation of constipation between 2015 and 2021. Using age and sex‐specific normal values, HR‐ARM recordings were re‐analyzed and abnormalities were reported using the London Classification. A combination of Wilcoxon rank sum and Fisher's exact test were used.

Results

36 patients (19 women) with median age 71 (interquartile range [IQR]: 69–74) years, were included. Using the London Classification, 7 (19%) patients had anal hypotension, 17 (47%) had anal hypocontractility, and 3 women had combined hypotension and hypocontractility. Anal hypocontractility was significantly more common in women compared to men. Abnormal BET and dyssynergia were noted in 22 (61%) patients, while abnormal BET and poor propulsion were only seen in 2 (5%). Men had significantly more paradoxical anal contraction and higher residual anal pressures during simulated defecation, resulting in more negative recto‐anal pressure gradients. Rectal hyposensitivity was seen in nearly one third of PD patients and comparable among men and women.

Conclusion

Our data affirms the high prevalence of anorectal disorders in PD. Using the London Classification, abnormal expulsion and dyssynergia and anal hypocontractility were the most common findings in PD. Whether the high prevalence of anal hypocontractility in females is directly related to PD or other confounding factors will require further research.

Gastrointestinal (GI) dysfunction is one of the most common nonmotor symptoms in Parkinson's Disease (PD), occurring in about 60–80% of patients with PD. ¹ , ² , ³ These include constipation, defecatory disorders, dysphagia, gastroparesis, and small intestinal bacterial overgrowth. ⁴ Indeed GI tract is thought to have an important role in the pathophysiology of PD. ⁵ , ⁶ Among the various GI disturbances seen in PD, constipation remains one of the most prevalent symptoms, seen in 80% of PD patients. ² , ⁷ It is also one of the earliest non‐motor symptoms to occur in PD and is proposed as a factor to increase the risk of future PD. ⁸ , ⁹ , ¹⁰ Constipation is shown to be associated with cognitive decline in PD and its severity can determine the progression to dementia. ¹¹ , ¹²

Despite its high prevalence and prognostic value, constipation often goes unrecognized during the early stages of PD. As the disease progresses, the severity and frequency of constipation often worsen, thus causing a significant burden to patients and their caregivers. ¹³ Early recognition and optimal medical management of constipation not only can improve quality of life, but also lessen the prevalence of emergency visits and hospitalizations in such patients. ¹⁴ , ¹⁵

The etiology of constipation in patients with PD is likely multifactorial. Proposed hypotheses include slow transit constipation due to neuronal loss in the submucosal and myenteric plexi, infiltration of Lewy bodies in the vagus nerve. ² Pelvic floor dyssynergia with paradoxical anal sphincter and puborectalis contraction during defecation has been increasingly recognized as the underlying etiology for defecatory dysfunction induced constipation in PD. ³ , ¹⁶ Dyssynergia can be secondary to focal dystonia of pelvic floor muscles evident with involuntary paradoxical anal sphincter contraction during defecation with or without impaired rectal sensation due to afferent pathways involvement. ¹⁷ , ¹⁸

Anorectal physiologic tests such as high resolution anorectal manometry (HR‐ARM) and balloon expulsion tests (BET) are currently recommended as the initial steps in evaluation of chronic constipation. ¹⁹ Studies examining the HR‐ARM profiles across PD patients have variably identified dyssynergic defecation, hypercontractile anal sphincter and, impaired anal squeeze pressures. ¹⁶ , ¹⁷ , ¹⁸ , ²⁰ While these findings suggest a role for underlying defecatory dysfunction as the pathology of PD patients with constipation, the inconsistency in how HR‐ARM studies are performed and interpreted may affect decision‐making and outcomes. London Classification system proposed a standardized model for reporting and interpretation of HR‐ARM findings in 2020. ²¹ London classified anorectal pathologies into disorders of recto‐anal inhibitory reflex, anal tone and contractility, recto‐anal coordination and sensation.

Our study examined HR‐ARM profiles in women and men with PD‐associated defecatory dysfunction induced constipation and utilized the London Classification to standardize the findings. Specifically, we were interested in investigating any differences in anorectal function between women and men.

Materials and Methods

Patient Population

Electronic medical records of patients with PD who completed HR‐ARM between 2015 and 2021 at a tertiary care Gastroenterology clinic for evaluation of chronic constipation were reviewed. Patients were grouped based on their gender. HR‐ARM results were compared to age‐ and gender‐specific normal values. Demographic data was collected, along with HR‐ARM results, and relevant comorbidities and symptoms. The study was approved by the institutional review board. Prior to testing, patients were advised to discontinue any cholinergic agents, promotility agents, stool softeners, or laxatives for 48–72 hours.

High Resolution Anorectal Manometry (HR‐ARM)

Patients were instructed to administer one fleet enema 2 hours prior to the HR‐ARM procedure. The procedure was performed in the left lateral position with patients’ knees and hips flexed. A two‐dimensional solid‐state HR‐ARM catheter (Medtronic, Minneapolis, MN, USA) was inserted into the rectum. Rectal propulsive pressures were captured from the proximal sensor located in the rectal balloon, and the average of pressures from anal sensors through the e‐sleeve provided the anal pressures. Fig. 1 shows the rectal and anal pressures during defecation in a healthy subject as compared to dyssynergia. With the catheter in place, patients were instructed to relax, squeeze, bear down, and cough, according to a previously reported protocol. ²² Resting pressures were recorded after 3 to 5 minutes of acclimation to the catheter. Squeeze pressures were obtained three times with 1 minute of rest in between. Each time, the patient was asked to hold the squeeze for 20 seconds. Patients were then instructed to push and bear down to simulate evacuation for 20 seconds (with a deflated rectal balloon). This was repeated twice with a 1‐minute resting period in between. The recto‐anal inhibitory and cough reflexes were assessed.

FIG. 1.

Anorectal manometric patterns during simulated defecation in healthy controls versus dyssynergic defecation. Schematic view of anorectal manometry catheter with respect to the rectal sensors showing the pressures, top panels and anal sensors showing the anal pressures, the lower panels. (A) Normal anal relaxation during simulated defecation. (B) Paradoxical anal contraction during simulated defecation, consistent with dyssynergic defecation.

Rectal sensation testing was performed by distending a rectal balloon in 10 mL increments until the first sensory threshold was reached, and in 30 mL increments thereafter until a maximum volume of 350 mL was reached, or until the patient reported severe urgency (whichever was reached first). Analysis was done using the ManoScan Version 3 (Medtronic, Minneapolis, MN, USA). We compared the cohorts anorectal pressure profile with sex and age matched published normative values. ²³

Balloon Expulsion Test (BET)

Balloon expulsion was performed using a non‐latex rectal balloon (Mui Scientific, Mississauga, ON, Canada) filled with 50 mL of water. Patients were asked to expel the balloon while sitting on a bedside commode. If patients could not expel the balloon in 2 minutes, it was deflated and removed manually. We chose 2 minutes as the upper limit of normal. ²⁴ Although a cut off of 1 minute has been proposed for normal BET, maximum allowed expulsion time (between 1 and 5 minutes) did not appear to significantly affect test performance relative to reference tests based on a recent systemic review and meta‐analysis. ²⁵

Data Analysis

Characteristics collected included age, body mass index (BMI), race, comorbidities, and GI symptoms. HR‐ARM variables evaluated included sphincter pressures and rectal sensory levels as well as BET.

Using the London Classification, anorectal disorders were reported based on manometric patterns and compared between men and women with PD and constipation. Mean sphincter pressures were compared between men and women using Wilcoxon rank sum test. The categorical variables were compared using Fisher's exact test. All statistical analyses were done using Stata version 14 (StataCorp LLC, College Station, TX, USA). In all cases, P < 0.05 was considered significant.

Results

Patient Demographics

We reviewed 36 individuals with PD who underwent HR‐ARM for evaluation of constipation and compared findings between males and females. Median age of the entire cohort was 71, interquartile range (IQR); 69–74 years. Common comorbidities included gastroesophageal reflux disease (GERD), irritable bowel syndrome (IBS), and depression/anxiety which overall, were more commonly seen in women. Table 1 shows demographics, comorbidities and GI symptoms in each group.

TABLE 1.

Baseline demographics

Variable	All (n = 36)	Female (n = 19)	Male (n = 17)	p‐value
Basic Demographics
Age: Median (IQR)	71 (69–74)	74 (69.4–77.9)	71 (69.6–74)	0.210
BMI: Median (IQR)	25.4 (22.9–29.7)	22.7 (20.5–26.3)	25.5 (23–29.7)	0.039
BMI > 25, n (%)	15 (41)	9 (52)	6 (31)	0.311
Race, n (%)				0.158
White	24 (67)	15 (79)	9 (53)
Other	12 (33)	4 (21)	8 (47)
Comorbidities, n (%)
Depression/Anxiety	22 (61)	12 (63)	10 (59)	1.000
IBS	13 (36)	8 (42)	5 (29)	0.502
GERD	20 (56)	14 (74)	6 (35)	0.042
Narcotic Use	4 (11)	3 (16)	1 (6)	0.605
Gastrointestinal Symptoms, n (%)
Bloating	23 (64)	12 (63)	11 (65)	1.000
Abdominal pain	21 (58)	12 (63)	9 (52)	0.736
Diarrhea	11 (31)	7 (37)	4 (24)	0.481
Dyspepsia	12 (33)	7 (37)	5 (29)	0.732

Note: Bold Text indicates statistically significant (P < 0.05) differences.

Abbreviations: GERD, gastroesophageal reflux disease; IBS, irritable bowel syndrome.

Out of the 36 patients who underwent GI evaluation and HR‐ARM, 18 patients had HR‐ARM after being diagnosed with PD, 10 patients had HR‐ARM before being formally diagnosed with PD, and eight had received the diagnosis of PD within the same year as their HR‐ARM. The median (IQR) time between the diagnosis of PD and HR‐ARM was 4.4 (3–7.8) years for patients with known PD, and for those who had HR‐ARM before or around the same time of being diagnosed with PD, the median (IQR) time was −0.2 (−2.5–0.2) years. There was no significant correlation between anorectal pressures and the duration of PD diagnosis (P > 0.05).

At the time of HARM, 20 patients were not taking any medications for PD due to either lack of diagnosis or intolerance to medication side effects. On the other hand, the remaining 16 patients were taking Carbidopa–Levodopa. However, the rate of anorectal motor and sensory abnormalities between the patients who were taking medications and those who were not there was not significant (P > 0.05).

Anorectal Disorders Based on London Classification

Table 2 summarizes the reporting of anorectal disorders using the London Classification, according to which, they were categorized as disorders of: (1) anal tone and contractility, (2) recto‐anal coordination, and (3) rectal sensation. Of the 36 patients, 7 (19%) had anal hypotension, 17 (47%) with anal hypo‐contractility, and only 3 women had both. Anal hypo‐contractility was significantly more common in females compared to males (P = 0.01). While dyssynergia was seen in all except one man and one woman, abnormal expulsion with dyssynergia was noted in 22 (61%) of patients (12 females, 10 males). This was the most common disorder for recto‐anal coordination noted in this cohort. This disorder was significantly higher than abnormal expulsion with poor propulsion and combined propulsion with dyssynergia (both 2, 5% of patients). Major rectal sensation abnormalities with rectal hyposensitivity and hypersensitivity were reported in 12 (33%) and 3 (8%) patients respectively and were similar in both males and females.

TABLE 2.

London classification of manometric parameters for findings of anorectal disorders

Variable	All (n = 36)	Female (n = 19)	Male (n = 17)	p‐value
Disorders of anal tone and contractility
Combined hypotension & hypocontractility	3 (8)	3 (16)	0 (0)	0.23
Anal hypotension	7 (19)	4 (21)	3 (17)	1
Anal hypocontractility	17 (47)	13 (68)	4 (23)	0.01
Disorders of rectoanal coordination
Abnormal expulsion & poor propulsion	2 (5)	1 (5)	1 (6)	1
Abnormal expulsion & dyssynergia	22 (61)	10 (52)	12 (70)	0.32
Abnormal expulsion & poor propulsion & dyssynergia	2 (5)	1 (5)	1 (6)	1
Disorders of rectal sensation
Rectal hyposensitivity	12 (33)	7 (37)	5 (29)	0.73
Rectal hypersensitivity	3 (8)	2 (10)	1 (6)	1

Note: Bold Text indicates statistically significant (P < 0.05) differences.

HR‐ARM and BET Findings

Table 3 summarizes the data collected from HR‐ARM and BET. Maximum anal squeeze pressures (in mmHg) were significantly higher in males (162.7 vs. 89, P = 0.0005). Residual anal pressures (in mmHg) and paradoxical contraction during simulation defecation were also significantly higher in males (83.2 vs. 51.7, P = 0.0012; −24 vs. −6, P = 0.0393) respectively. A significantly higher recto‐anal pressure gradient was seen in males (−71.3 vs. −19.5, P = 0.0056). Sensory levels and balloon expulsion (in ml) were similar in both groups.

TABLE 3.

Anorectal pressure profile

Variables; median and interquartile range (IQR)	Female (n = 19)	Male (n = 17)	p‐value
Mean resting sphincter pressure (mm Hg)	42.7 (34.5–69.4)	58.2 (39.6–69.8)	0.1994
Max resting sphincter pressure (mm Hg)	45.7 (40–75.6)	67.2 (43.1–80)	0.1322
Max squeeze sphincter pressure (mm Hg)	89 (42.8–122.1)	162.7 (137.6–211.4)	0.0005
Duration of sustained squeeze (s)	16.5 (4.9–20.5)	9.9 (6.3–19.2)	0.5788
Simulated defecation
Residual anal pressures (mm Hg)	51.7 (41.1–75.4)	83.2 (68.1–110.4)	0.0012
% Relaxation	‐6 (−17–8)	−24 (−60–9)	0.0393
Rectal pressure (mm Hg)	32.5 (23.1–43.8)	31.8 (24.9–45.5)	0.7393
Rectoanal pressure gradient	−19.5 (−37.7–6.1)	−71.3 (−74.9–75)	0.0056
Sensory levels
First sensation (mL)	30 (20–60)	40 (20–70)	0.6885
Urge to defecate (mL)	70 (60–120)	80 (60–120)	0.5717
Maximum tolerated volume (mL)	125 (80–160)	130 (100–210)	0.3642
Balloon expulsion test, n (%)
Abnormal expulsion	10 (55)	13 (76)	0.289

Discussion

Constipation and defecatory dysfunction are prevalent in patients with PD. ¹⁷ , ¹⁸ This is the first study using the London Classification‐ to uniformly interpret and report HR‐ARM findings in men and women with PD. Using the London Classification, we have presented the prevalence of anorectal disorders in PD patients in a more structured and standardized format. In doing so, we have demonstrated how PD affects all aspects of anorectal function, which include abnormalities in anal tone/contractility, recto‐anal coordination, and rectal sensation. Using this classification allows physicians to better understand anorectal pathology in PD patients in a standardized fashion across institutions, which can yield more targeted medical management and therapy. From the research perspective, this approach will facilitate the proper definition of endpoints of anorectal neuromuscular dysfunction and their link to clinical outcomes in patients with PD.

Obtaining a detailed bowel history will help assessment of patients with PD who are experiencing constipation and defecatory dysfunction. Infrequent bowel movements with or without hard stools are the most commonly recognized symptoms of constipation by clinicians. Other symptoms such as the need to strain during bowel movements, feeling of blockage in the rectum, sensation of incomplete defecation, and the requirement for digital manipulation to facilitate rectal evacuation also indicate constipation. ¹⁹ While these symptoms may suggest presence of defecatory dysfunction, anorectal physiologic tests such as anorectal manometry and BETs are required for precise diagnosis. ²⁶ Recognition of the cause of constipation plays a significant role in clinical decision‐making. The management of defecatory dysfunction‐induced constipation involves pelvic floor physical therapy and biofeedback, whereas the use of laxatives has been found to be mostly ineffective. ²⁷ , ²⁸ Nevertheless, in PD patients with overlapping defecatory dysfunction and slow transit constipation, a combination of laxatives, physical therapy, and bowel habit training may be necessary. We found that disorders of recto‐anal coordination with combined abnormal BET and dyssynergia was the most common finding seen in 61% of PD patients. Our findings are in line with the findings of Triadafilopoulos et al. where 88% of 58 PD patients had abnormal balloon expulsion. ¹⁶ A high prevalence of dyssynergia was also noted in a Perugia Clinic cohort of 31 PD patients where 19 out of 31 patients were noted to have HR‐ARM findings consistent with dyssynergia. ¹⁸ Our findings reaffirm that a key contributing factor to constipation in PD patients is defecatory dysfunction, as demonstrated by prior studies. ⁴ Additionally, we found diminished anal tone in 20% of this cohort while anal hypocontractility was seen in nearly half of PD patients. In PD, fecal incontinence is a significant issue. As per our prior study, 24% of individuals experienced passive fecal incontinence and 76% experienced urge incontinence, which could be linked to anal hypotonicity and/or overflow incontinence surrounding fecal impaction. ¹⁶ Decreased anal resting tone has been reported at much higher rate (58%) along with diminished anal squeeze pressures in (76%) in another study of HR‐ARM. The observed differences are likely to be related to the strict criteria we used for sex and age matched normative values. ²⁹ A high prevalence of anal hypo‐contractility was also noted in a Nanjing clinic cohort of 35 PD patients age and sex matched to 35 patients with functional defecation disorder who underwent HR‐ARM and noted lower maximum anal squeeze pressures in patients with PD compared to other patients with defecatory dysfunction (196.3 vs. 232.6 mmHg, P = 0.011). ³⁰ The etiology of reduced squeeze pressures in PD patients is likely multifactorial, stemming from pelvic floor dystonia, and Lewy body infiltration in the nerves and muscles innervating the sphincter muscles, resulting in reduced squeeze pressures. ³⁰ The prevalence of abnormal anal tone and contractility could explain the concurrent symptoms of fecal incontinence that is often reported in PD patients. Nevertheless, normal rectal sensory function is essential for proper defecation and maintenance of continence. We showed high rate of abnormal rectal sensation in nearly half of PD patients (hyposensitivity in one third and rectal hypersensitivity in another 8%) which is consistent with previous reports. ¹⁶ The rectal sensory abnormalities indicate parasympathetic afferent dysfunction in the setting of dysautonomia. Altogether, HR‐ARM demonstrated multiple anorectal pathologies involved in PD of which some are likely to indicate autonomic dysfunction as well. However, retrospective studies have failed to show a correlation between severity of PD based on validated questionnaires or disease duration and anorectal pressure profile or the colonic transit time. ¹⁶

We further found that, while disorders of recto‐anal coordination and rectal sensation were similar in men and women, disorders of anal tone/contractility, and specifically, anal hypo‐contractility, were significantly more common in women than men with PD. When comparing specific HR‐ARM parameters between women and men with PD, almost all values were lower in women compared to men as expected in healthy controls. ²³ Maximum squeeze sphincter pressures were significantly diminished in women. While there is a significant difference noted between women and men, it is unclear whether additional confounding factors (such as multiparity, labor and delivery, prior anal surgery) may be contributing to this difference. Prior studies have demonstrated that, in general, maximum squeeze sphincter pressures are lower in women compared to age‐matched men due to several reasons including: further developed and thicker sphincters in males due to testosterone effect, and history of pregnancy and vaginal deliveries which involves traction injury to the pudendal nerve, which innervates the sphincter muscles. ²⁴ , ³¹ , ³² However, our study demonstrates that squeeze pressures in both men and women with PD were generally lower than the age‐ and sex‐matched healthy controls, suggesting that the pathophysiology of PD may have a greater significant effect on the anatomical and functional aspect of defecation.

Our study also confirms, that, as in prior studies, significant differences seen in the HR‐ARM parameters during simulated defecation between genders. Anal relaxation was significantly less in men as compared to women with PD (−24% vs. −6%) as opposed to other patient populations with chronic constipation which showed only slight difference between genders (6% vs. 6.6%). ³³ Given comparable rectal pressures among men and women the significant differences in the anal relaxation resulted in a significantly more negative recto‐anal pressure gradients in men compared to women. These findings may suggest that the paradoxical anal contraction in patients with PD may be related to focal dystonia and therefore more pronounced compared to other patients with defecatory disorders.

The findings of our study allow better understanding of the pathology of chronic constipation in patients with PD, and demonstrates how HR‐ARM can be a valuable diagnostic tool to better discern the potential etiologies of refractory constipation in PD. Nevertheless, addressing constipation is especially important given it being a prognostic factor associated with overall lower quality of life in PD. These findings can help educate and guide neurologists and, in turn, can help improve better outcomes with use of alternative therapies that they might not have been considered otherwise. Currently, patients with PD and constipation are advised to frequent use of laxatives and enemas in conjunction with hydration and high‐fiber intake without addressing the anorectal pathology that is often present. Our study highlights that anorectal pathology plays a critical role on the burden of constipation in PD patients. As such, this brings awareness to the importance of managing these disorders with the use of targeted therapies such as pelvic floor therapy, biofeedback, and botox injection to the pelvic floor. ³⁴

Our study is limited by its retrospective nature, and predisposition to selection and referral biases, as it was conducted at a tertiary referral center. To minimize errors, we used age‐ and sex‐specific normal values as baseline comparisons when interpreting HR‐ARM patterns. Lack of validated questioners to correlate the severity of PD motor symptoms with constipation severity and anorectal pressures is another limitation of this study. In our previous work, we noted that the usage of HR‐ARM and wireless motility capsule can transform the non‐specific symptoms of constipation and defecatory dysfunction into distinct disorders, such as dyssynergia, slow transit constipation and overlap syndromes and may guide specific clinical management in patients with PD. Of note, previous research did not show any correlation between PD scores and disease duration with either the manometric or colonic transit findings. ¹⁶ Although in this study the duration of PD and PD medications did not seem to impact the anorectal pressure profiles, understanding the impact of PD duration, severity, and PD therapy on GI function is pivotal and needs to be addressed in a prospective long‐term study wherein Unified Parkinson’s Disease Rating Scale (UPDRS)/Movement Disorder Society (MDS)‐UPDRS III or at least the Hoehn and Yahr stage obtained in parallel with HR‐ARM data. Such study will have to be conducted over several years since the HR‐ARM findings do not change over short intervals. Further, as in our cohort, constipation often precedes the diagnosis of PD by many years, complicating the relationship between symptoms and objective findings. ³⁵ The limited number of patients in our study would not allow for meaningful conclusions on identifying relationship between HR‐ARM and other non‐motor symptoms. However, such effort is hampered by the fact that other concomitant issues, such as dysautonomia, may have an effect on the HR‐ARM findings. ³⁶ In this study we did not assess for anorectal structural pathologies or colonic transit yet showed that anorectal dysfunction appears to be a key driver in constipation of PD patients. Nonetheless, 86% of PD patients had abnormal HRAM contributing to their constipation. Hence the anorectal dysfunction plays a major role, and its proper management would be more valuable in quality of life (QoL) improvement. Future studies are needed to show the burden of such pathologies on constipation symptom burden in PD patients.

In summary, using the London Classification, we have confirmed a high prevalence of various anorectal disorders between men and women with PD. We have highlighted that abnormal expulsion with dyssynergia and anal hypo‐contractility are the most common anorectal disorders in PD. Some key HR‐ARM parameters, particularly anal contractility and dyssynergia are different across women and men with PD and should be further examined. When confronted with such PD patients with constipation and defecatory difficulties, movement disorder specialists should consider further objective evaluation using HR‐ARM and plan individualized therapy for the management of symptoms, based on the information obtained, instead of empirical therapies that may not only be ineffective, but also possibly counterproductive. Future research on a larger population size comparing varying stages of PD (early, mid, advanced) and the differences in HR‐ARM profiles between men and women using the London Classification can help guide more targeted therapy of pelvic floor disorders in PD.

Author Roles

(1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique (3) Manuscript: A. Writing of the first draft, B. Review and Critique.

W.Z.:1C, 2C, 3A, 3B

G.T.: 1A, 1B, 2C, 3B

B.G.: 3B

H.H.: 1A, 1B, 2C, 3B

L.B.: 3B

P.G.: 3B

L.N: 3B

M.M.: 3B

S.M.: 3B

D.S.: 3B

S.J.: 3B

L.N.: 1A, 1B, 1C, 2A, 2B, 2C, 3B

Disclosures

Ethical Compliance Statement: Our study was approved by the Stanford University School of Medicine institutional review board (#51596). This was a retrospective cohort study and patient consent was not required for data collection or analysis, as approved by the aforementioned institutional review board. The authors confirm that patient consent was not required for this work. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflicts of Interest: Study was not funded and there is no conflict of interest.

Financial Disclosures for the Previous 12 Months: The authors declare that they have no conflicts of interest.