Impact of referral delays on dysphagia outcomes in a Saudi tertiary center

Abstract

BACKGROUND

Dysphagia requires prompt evaluation but data regarding referral-to-assessment intervals and its association with outcomes in tertiary care are scarce, especially in Saudi Arabia.

AIM

To investigate the assessment and outcome of consecutive dysphagia referrals to a tertiary gastroenterology practice.

METHODS

This retrospective single-center study analyzed 207 consecutive dysphagia referrals to King Saud Medical City (2022-2023). Data collection included referral timelines, diagnostic investigations, etiologies and outcomes. Statistical Package for the Social Sciences version 28.0, employing χ² tests and analysis of variance, with P < 0.05 considered statistically significant.

RESULTS

Total 168 patients included in this study (mean age 45.7 ± 17.7 years, 50% male), referral delays > 2 weeks occurred in 44% for clinical assessment and 50% for endoscopy. Gastroesophageal reflux disease was most common (45.2%), followed by eosinophilic esophagitis (14.8%) and malignancy (6.5%). Patients receiving endoscopy within 2 weeks showed an 84.6% improvement, compared to 76.0% with delayed referral (P = 0.012). All malignant cases were referred within 2 weeks, compared to 52.7% of non-malignant cases (P = 0.013). However, 67% of the malignant cases worsened, and 33% died.

CONCLUSION

Early endoscopy within 2 weeks provides significant benefit. Optimised management of dysphagia should consist of more direct referral pathways.

Key Words: Dysphagia; Referral timelines; Endoscopy; Gastroesophageal reflux disease; Malignancy; Tertiary care; Saudi Arabia

Core Tip: This is a retrospective analysis of 168 dysphagia patients, which shows that delays > 2 weeks are common (44%-50%), but early endoscopy significantly improves outcomes (84.6% vs 76.0% improvement, P = 0.012). Despite prioritised referral of malignant cases, the prognosis was dismal which underscores the importance of earlier patient presentation. Direct endoscopy within 2 weeks, mainly prioritising streamlined pathways for alarm features, could be a more cost-effective approach in management and can lead to better survival rates.

INTRODUCTION

Most gastroenterology care requires referral. Dysphagia is a condition affecting approximately 3% of adults[1]. Oropharyngeal dysphagia defined as abnormality in initiating swallowing and coordinating the movement of solids and liquid from mouth to pharynx whereas the difficulty of swallowing from pharynx to gastroesophageal junction is esophageal dysphagia[2,3]. Timely access and assessment are associated with improved disease outcomes and patient satisfaction[4].

Dysphagia indicates a need for referral to gastroenterology care[5]. Multiple etiologies of dysphagia range from structural, inflammatory, and motility, with the most concerning being malignant neoplasm. However, the most common causes are benign in origin[6,7]. However, even in these situations, complication of weight loss and malnutrition persists[8].

Dysphagia is a prevalent presentation to Saudi tertiary centers, but little is known about local referral patterns and their impact on clinical outcomes. A lack of understanding on this issue makes it hard to know if current referral timings enable timely diagnosis and appropriate management. When it comes to a complete assessment of dysphagia, advanced diagnostic modalities are necessary. The diagnostic work-up of dysphagia usually involves a combination of endoscopy, esophageal manometry, barium swallow studies and/or computerized tomography[9]. Chicago classification criteria define achalasia subtypes as type I (absent peristalsis with minimal pressurization), type II (absent peristalsis with panesophageal pressurization), and type III (absent peristalsis with spastic contractions) and essentially rely on high-resolution manometry to confirm achalasia diagnosis and distinguish others motility disorders[5]. The pH-impedance may be used to identify the concomitant gastroesophageal reflux disease (GERD). Systematic endoscopic assessment allows for proper grading of gastroesophageal reflux disease (GERD) severity using Los Angeles (LA) classification as follows: (1) Grade A (mucosal breaks < 5 mm); (2) Grade B (5-10 mm); (3) Grade C (> 10 mm, < 75% circumference); and (4) Grade D (> 75% circumference).

Clinical guidelines vary significantly in their recommendations regarding when an investigation should be conducted following dysphagia referrals. The British Society of Gastroenterology’s guidelines recommend urgent upper endoscopy within two weeks for all dysphagia patients[5], while Canadian consensus guidelines suggest endoscopy within two months for non-urgent cases while reserving expedited evaluation for severe or rapidly progressive dysphagia[10]. However, both recommendations are based on limited evidence.

This study aims to: (1) Audit adherence to international consensus guidelines and identify adverse outcomes related to delays; and (2) Determine whether referral processes can be improved to identify patients requiring expedited access. Our overall goal is to retrospectively analyze dysphagia management in our tertiary center, specifically evaluating the impact of referral timelines on diagnostic and clinical outcomes.

MATERIALS AND METHODS

Study design

This retrospective single-center study was conducted at King Saud Medical City, a tertiary referral center in Riyadh, Saudi Arabia, over 2 years (2022-2023).

Patient population

Total 207 consecutive dysphagia referrals were identified through Tanseeq and Watheeq electronic referral systems. A patient flow chart summarizing selection criteria and exclusions is presented in Figure 1. Chart reviews were performed for all referrals, identifying: (1) The patient’s age, biological sex; (2) Previously identified etiology of dysphagia; (3) Medications that may impact dysphagia (opioids, benzodiazepines, antiepileptic drugs, and antipsychotics); (4) Presenting features (frequency of dysphagia, progressive dysphagia, weight loss); (5) Presence of typical reflux symptoms (heartburn and regurgitation); (6) Time from referral to assessment in weeks; (7) Time from referral to esophageal investigation in weeks; (8) Time from clinical assessment to esophageal investigation in weeks; (9) Etiology determined to be non-esophageal and referral made for non-esophageal investigations; (10) Etiology of esophageal dysphagia; (11) Treatment of esophageal dysphagia; (12) Complications related to etiology of dysphagia; (13) Complications related to treatment of dysphagia; and (14) Outcome (worsening, no change, improved).

Figure 1 Patient flow chart showing selection criteria, exclusions, diagnostic workup, and outcomes for dysphagia referrals to King Saud Medical City gastroenterology department (2022-2023). GERD: Gastroesophageal reflux disease.

All patients aged 18 years or older referred for dysphagia were included. Patients aged < 18 years (n = 9), or who did not attend their endoscopy appointment, refused upper endoscopy (n = 2), or were lost to follow-up without a final diagnosis (n = 28), were excluded. Charts were reviewed based on the patient's medical record number (MR). All patients were given a study number that corresponds to the MR number. This information was stored in a separate encrypted file. All data were collected into a database that only identified patients by study number. This database did not contain any personal identifiers and was stored on an encrypted, password-protected memory storage device after analysis was complete to ensure confidentiality. Data in these computerized files was encrypted, and the encryption key was kept by the principal investigator. The study data were available only to the research team. Participants were not identified in any publication or reports.

Variable definitions

Dysphagia classification: Oropharyngeal dysphagia was diagnosed based on clinical presentation, swallowing evaluation, and neurological assessment. Esophageal dysphagia was confirmed by endoscopy, imaging studies, and manometry when indicated.

GERD severity assessment: GERD diagnosis was based on the endoscopist's final clinical impression of LA classification.

Achalasia classification: When high-resolution manometry was available, achalasia was classified according to Chicago classification criteria into types I, II, and III based on peristaltic patterns and pressurization characteristics. Due to this limitation, achalasia was diagnosed based on clinical symptoms and barium swallow.

Manometry and 24-hour pH: Due to the newly established nature of our motility unit during the study period, studies were performed selectively, based on equipment availability and clinical indications, during our unit's development phase. Patient improvement was assessed based on subjective symptom improvement reported at follow-up visits, reflecting real-world clinical practice patterns.

Ethical consideration

The study was approved by the Institutional Research Ethics Board at King Saud Medical City (No. H1RI-03-Jun24-01). Patient confidentiality and anonymity were maintained throughout the process.

Statistical analysis

Data were entered and analyzed using Statistical Package for the Social Sciences 28.0. Descriptive statistics included frequencies and percentages for categorical variables; as well as means, ranges, and SD for continuous numerical variables. The χ² or Fisher's exact test was used to assess associations between categorical variables. One-way analysis of variance (ANOVA) test was used to compare means of a continuous variable between groups, while post-hoc Tukey test was used in case of significant ANOVA to compare between individual groups. P < 0.05 was considered statistically significant. An audit was conducted to assess how well the local metrics align with international consensus guidelines, e.g., Canadian[10], British[5], and United States[11] recommendations.

RESULTS

Baseline characteristics

A total of 168 patients were included in the study following the patient flow outlined in Figure 1. Their baseline characteristics are summarized in Table 1. The study population was equally distributed by gender (82 males, 82 females), with ages ranging from 18 years to 87 years (mean 45.7 ± 17.7 years). Most patients were referred by the family medicine department (69.4%). Previous dysphagia etiology was reported in 9.9% of patients; mainly corrosive/caustic ingestion, esophageal cancer, and central nervous system glitches (21.4% for each). Dysphagia duration ≥ 4 weeks occurred in 44.5% of cases. Progressive dysphagia was present in 48.5% of patients. Two-thirds (66.7%) experienced retrosternal (mid-chest) dysphagia, while one-third (33.3%) experienced cervical (neck-level) dysphagia. Associated symptoms included heartburn (64.6%), regurgitation (58.0%), weight loss (42.2%), chest pain (35.9%), and globus sensation (23.4%).

Table 1 Baseline characteristics and clinical features of patients with dysphagia referred to King Saud Medical City (2022-2023), n (%).

Baseline characteristics	Total study population (n = 168)	Early endoscopy (< 2 weeks) (n = 36)	Delayed endoscopy (≥ 2 weeks) (n = 45)	P value
Demographics
Age (years), mean ± SD	45.7 ± 17.7	48.0 ± 17.7	44.2 ± 16.7	0.365
Male gender	82 (50.0)	19 (52.8)	18 (40.0)	0.356
Referral source
Family medicine	116 (69.4)	19 (52.8)	38 (84.4)	0.004
Emergency department	19 (11.4)	7 (19.4)	0 (0.0)	0.002
General surgery	12 (7.2)	5 (13.9)	1 (2.2)	0.083
Otorhinolaryngology	7 (4.2)	2 (5.6)	3 (6.7)	1
Oncology	6 (3.6)	1 (2.8)	0 (0.0)	0.444
Infectious diseases	2 (1.2)	1 (2.8)	0 (0.0)	0.444
Clinical features
Duration ≥ 4 weeks	57 (44.5)	18 (50.0)	22 (48.9)	1
Progressive dysphagia	64 (48.5)	15 (41.7)	19 (42.2)	1
Retrosternal location	86 (66.7)	28 (84.8)	25 (58.1)	0.024
Weight loss	54 (42.2)	14 (38.9)	20 (44.4)	0.782
Heartburn	84 (64.6)	23 (63.9)	29 (64.4)	1
Regurgitation	76 (58.0)	21 (58.3)	25 (55.6)	0.98
Chest pain	47 (35.9)	12 (33.3)	16 (35.6)	1
Globus sensation	30 (23.4)	3 (8.3)	6 (13.3)	0.724
Diagnostic workup
Manometry performed	9 (8.5)1	3 (8.3)	4 (8.9)	1
24-hour pH study	3 (2.3)1	1 (2.8)	1 (2.2)	1
Esophagogram performed	65 (38.7)	12 (33.3)	18 (40.0)	0.647
Final diagnoses
GERD	76 (45.2)	14 (38.9)	20 (44.4)	0.782
Eosinophilic esophagitis	25 (14.8)	0 (0.0)	6 (13.3)	0.031
Complicated GERD with stricture	9 (5.4)	8 (22.2)	1 (2.2)	0.009
Malignancy	11 (6.5)	1 (2.8)	0 (0.0)	0.444
Achalasia	10 (5.9)	1 (2.8)	5 (11.1)	0.219
Other diagnoses	37 (22.0)	12 (33.3)	13 (28.9)	0.809
Clinical outcomes
Improved outcome	103 (61.3)	23 (63.9)	27 (60.0)	0.898

¹Percentage of 106 patients who underwent evaluation.

GERD: Gastroesophageal reflux disease.

Referral time

Referral delays exceeded two weeks across multiple care stages. Time from primary referral to clinical assessment by the gastroenterology department exceeded two weeks in 44% of patients, while referral to endoscopy exceeded two weeks in 50% of cases. Additionally, the time from primary site referral to undergoing esophagogram exceeded two weeks in 50.8% of patients. Between clinical assessment and endoscopy, delays exceeded two weeks in 27.2% of patients, while delays between clinical evaluation and performance of esophagogram exceeded two weeks in 43.7% of patients.

Diagnostic workup

Advanced diagnostic testing was limited during the establishment phase of our unit. Manometry was performed in 9 patients (8.5% of 106 evaluated), while 24-hour pH studies were conducted in 3 patients (2.3%). Most endoscopic biopsies were obtained from the distal esophagus (61.5%), followed by the proximal esophagus. Among patients who underwent endoscopic evaluation, LA classification revealed grade A esophagitis in 3 patients (3.9%), grade B in 55 patients (72.4%), grade C in 9 patients (11.8%), and grade D in 9 patients (11.8%), with grade B representing the majority of cases (72.4%).

Etiology of dysphagia

GERD was the most common condition (45.2%), followed by eosinophilic esophagitis (14.8%), while malignancy was reported in 6.5% of cases (Figure 2A). Significant gender differences were observed: Eosinophilic esophagitis was more prevalent in males (23.4% vs 6.7%, P = 0.004), while GERD showed a female predominance trend (53.3% vs 39.0%, P = 0.076). Age distributions varied significantly by etiology (P < 0.001). Malignancy patients were older (63.8 ± 15.1 years) compared to eosinophilic esophagitis patients (33.7 ± 9.7 years, P < 0.001). Age of patients with oropharyngeal dysphagia was also significantly higher than those with eosinophilic oesophagitis (57.4 ± 21.5 vs 33.7 ± 9.7), P = 0.033.

Figure 2 Etiology and outcome of cases of dysphagia, King Saud Medical City, Riyadh (2022-2023). A: Etiology of dysphagia; B: Outcome of cases. GERD: Gastroesophageal reflux disease.

Clinical outcome

Most dysphagia patients (61.3%) improved after management, while 4.2% worsened and 1.8% died (Figure 2B). Early endoscopy significantly improved outcomes, as 84.6% of patients receiving endoscopy within two weeks showed improvement, compared to 76.0% with delays exceeding four weeks (P = 0.012) (Table 2).

Table 2 Association between the time of referral of cases presented with dysphagia and the outcome of management, n (%).

	Final outcome					P value1
	The same (n = 19)	Improved (n = 103)	Worsening (n = 7)	Wrong referral (n = 4)	Died (n = 3)
Time of referral to clinical assessment by gastroenterology department in weeks (n = 133)	n = 18	n = 0	n = 0	n = 3	n = 2
≤ 2 hours (n = 72)	6 (8.3)	57 (79.2)	6 (8.3)	1 (1.4)	2 (2.8)
3-hour (n = 22)	4 (18.2)	18 (81.8)	0 (0.0)	0 (0.0)	0 (0.0)
> 4 hours (n = 39)	8 (20.5)	28 (71.8)	1 (2.6)	2 (5.1)	0 (0.0)	0.246
Time from referral to endoscopy in weeks (n = 132)	n = 18	n = 0	n = 0	n = 3	n = 1
≤ 2 hours (n = 65)	3 (4.6)	55 (84.6)	6 (9.2)	0 (0.0)	1 (1.5)
3-hour (n = 17)	6 (35.3)	10 (58.8)	0 (0.0)	1 (5.9)	0 (0.0)
> 4 hours (n = 50)	9 (18.0)	38 (76.0)	1 (2.0)	2 (4.0)	0 (0.0)	0.012
Time from referral to esophagogram in weeks (n = 60)	n = 11	n = 47	n = 0	n = 2	n = 0
≤ 2 hours (n = 30)	3 (10.0)	26 (86.7)	0 (0.0)	1 (3.3)	0 (0.0)
3-hour (n = 6)	3 (50.0)	3 (50.0)	0 (0.0)	0 (0.0)	0 (0.0)
> 4 hours (n = 24)	5 (20.8)	18 (75.0)	0 (0.0)	1 (4.2)	0 (0.0)	0.226
Time from clinical assessment to endoscopy in weeks (n = 129)	n = 18	n = 100	n = 7	n = 3	n = 1
≤ 2 hours (n = 95)	9 (9.5)	77 (81.1)	6 (6.3)	2 (2.1)	1 (1.1)
3-hour (n = 14)	6 (42.9)	6 (42.9)	1 (7.1)	1 (7.1)	0 (0.0)
> 4 hours (n = 20)	3 (15.0)	17 (85.0)	0 (0.0)	0 (0.0)	0 (0.0)	0.046
Time from clinical assessment to esophagogram in weeks (n = 50)	n = 10	n = 39	n = 0	n = 1	n = 0
≤ 2 hours (n = 29)	3 (10.3)	26 (89.7)	0 (0.0)	0 (0.0)	0 (0.0)
3-hour (n = 8)	4 (50.0)	4 (50.0)	0 (0.0)	0 (0.0)	0 (0.0)
> 4 hours (n = 13)	3 (23.1)	9 (69.2)	0 (0.0)	1 (7.7)	0 (0.0)	0.054

¹χ² test.

Total 32 cases lost to follow-up were removed from the analysis.

All malignant cases received expedited referrals within two weeks (100% vs 52.7% non-malignant, P = 0.013). Despite rapid referral, prognosis remained poor: 67% of malignant cases worsened, and 33% died compared to 2.3% worsening and 0.8% mortality in non-malignant cases (P < 0.001) (Table 3).

Table 3 Comparison of malignant and non-malignant cases of dysphagia regarding the outcome of management, n (%).

	Non-malignant cases (n = 128)	Malignant cases (n = 6)	P value1
The same (n = 19)	19 (14.8)	0 (0.0)	< 0.001
Improved (n = 101)	101 (79.0)	0 (0.0)
Worsening (n = 7)	2 (2.3)	4 (66.7)
Wrong referral (n = 4)	4 (3.1)	0 (0.0)
Died (n = 3)	1 (0.8)	2 (100)

¹χ² test.

DISCUSSION

The impact of the time from referral to clinical assessment in gastroenterology, on diagnose and prognosis of patients with dysphagia was not investigated extensively. Thus, this study is a trial to investigate this issue in a highly qualified center in Riyadh, Saudi Arabia.

In the current study, the most frequently reported cause of dysphagia was GERD (45.2%), followed by esinophilic esophagitis (14.8%) whereas malignancy was reported in 6.5% of cases. Also, GERD was the most common cause of dysphagia reported among United States patients[11]. It has been reported that esophageal dysphagia is commonly caused by GERD and functional esophageal disorders, while eosinophilic esophagitis is triggered by food allergens; it has also been recommended that esophageal biopsies be done to ensure the diagnosis is accurate[12].

Our study demonstrates benefits of expedited endoscopy in the management of dysphagia. Of the patients receiving endoscopy within two weeks, 84.6% showed improved outcomes compared to 76.0% with delayed referral (P = 0.012). Our analysis reveals patterns in referral prioritization – specifically, emergency department referrals consistently received expedited endoscopy within two weeks (19.4% vs 0%, P = 0.002), whereas family medicine referrals were more likely to experience delays (84.4% vs 52.8% in the delayed group, P = 0.004). Notably, patients with retrosternal dysphagia were more likely to receive early endoscopy (84.8% vs 58.1%, P = 0.024), and complicated GERD cases were appropriately prioritized for evaluation (22.2% vs 2.2% in the early group, P = 0.009).

In this study, 44% of patients with dysphagia experienced a referral time of more than two weeks from the primary site to clinical assessment by the gastroenterology department, while Canadian guidelines recommend endoscopy within two months for non-severe dysphagia[10], and British guidelines advocate urgent endoscopy within two weeks for all dysphagia cases[5]. Our findings highlight discrepancies in meeting international benchmarks, particularly for high-risk populations. Similarly, the referral time to endoscopy exceeded two weeks in 50% of cases, and the time to esophagogram performance exceeded two weeks in 50.8% of cases. Additionally, the time between clinical assessment and endoscopy exceeded two weeks for 27.2% of patients, while the time between clinical evaluation and esophagogram performance exceeded two weeks for 43.7% of patients. Current guidelines recommend esophagogastroduodenoscopy to exclude esophageal etiologies in patients presenting with dysphagia[13]. Supporting this approach, Waters et al[14] recently emphasized that adults with new-onset dysphagia require urgent upper gastrointestinal endoscopy within two weeks to facilitate timely diagnosis and management.

This study revealed that 61.3% of the patients with dysphagia improved after management while minority (4.2%) got worse and 1.8% died. Many factors could impact the prognosis of cases of dysphagia; among these factors the etiology of dysphagia and the time of referral[15]. In the present study, most cases referred from the primary department of presentation or from clinical assessment to endoscopy performance in a short period (within two weeks) was associated with better outcome of management compared to those who referred later. This finding supports the importance of early referral of cases of dysphagia for further care on the disease prognosis.

This study showed that all malignant cases were referred to clinical assessment by gastroenterology department and then to endoscopy performance within two weeks. Despite of that, two-thirds of malignant cases got worse after management and the remaining third died. It has been documented that despite advances in diagnosis and therapy, the prognosis of malignant cases presented with dysphagia remains poor, with a less than 20% 5-year survival rate[16]. Furthermore, Allum et al[17] reported that 50% of malignant cases of dysphagia had either a locally advanced disease or distant metastasis on presentation, which negatively impacts long-term remission and cure rates.

In the present study, GERD was reported as a cause of dysphagia more in female patients than male patients; however, the difference did not reach a statistically significant level. In earlier studies conducted in Germany[18] and Australia[19], no gender difference was observed. However, the male predominance in eosinophilic esophagitis (23.4% vs 6.7% in females, P = 0.004) aligns with the international literature[20]. Interestingly, eosinophilic esophagitis was significantly more common in the delayed endoscopy group (13.3% vs 0%, P = 0.031), suggesting potential diagnostic opportunities with earlier evaluation.

In the present study, history of performing manometry or 24 hours pH study was found in only 8.5% and 2.3% of patients presented with dysphagia, respectively due to equipment limitations during the unit’s development. This represents both a limitation and real-world challenges faced by developing healthcare units. The British Society of Gastroenterology guidelines recommended performing high-resolution manometry when investigating dysphagia cases, particularly those with achalasia, as well as pH studies for patients with symptoms of GERD, patients not responding to high-dose proton pump inhibitors, and patients with symptoms of GERD responsive to proton pump inhibitors for whom surgery is planned[5].

Low public awareness of dysphagia as a symptom of malignancy in Saudi Arabia may contribute to delayed patient presentation. Early detection improves survival and reduces healthcare costs[21,22].

In Saudi Arabia, integrating dysphagia awareness into national non-communicable disease screening programs – alongside diabetes or hypertension screenings, for example – could enhance early diagnosis and align with vision 2030 goals for preventive healthcare[23]. Our findings support advocating for: (1) Standardized referral protocols prioritizing endoscopy within two weeks; (2) Streamlined pathways for patients with alarm features; (3) The enhanced utilization of advanced diagnostic modalities as equipment becomes available; and (4) Systematic follow-up protocols.

Some noteworthy limitations of the present study should be acknowledged. First our study was conducted during the establishment phase of our motility unit. Outcome assessments relied on subjective patient-reported symptom improvement rather than validated instruments.

Furthermore, comparative studies are rare, which may affect the quality of our discussion of the results. Despite these limitations, the present study is unique in our region, utilized an adequate sample size, and includes a comprehensive description of different referral times for patients presenting with dysphagia, which could be of use to practitioners. While our study evaluated institutional referral timelines, we did not assess the duration between symptom onset and initial patient consultation, which may further explain the poor outcomes in malignancies. Finally, the absence of patient-reported symptom onset timing and delay causes represents a limitation. Future studies should incorporate patient-reported delays to better explain barriers to early presentation.

CONCLUSION

In conclusion, the referral and diagnostic process for dysphagia often exceeded two weeks in multiple stages, including referral to clinical assessment, endoscopy, and esophagogram. Nearly a quarter of patients faced delays of over two weeks between clinical assessment and endoscopy. The most frequently reported cause of dysphagia was GERD.

All malignant cases were referred from the primary department of presentation to clinical assessment by gastroenterology department and then to endoscopy performance within two weeks. Despite of that, prognosis was poor. Suggesting this is due to advanced stage rather than due to timing of referral. Based on these results, we recommend enhancing referral of patients presented with dysphagia directly to endoscopy to improve the prognosis. Further longitudinal multicentric study is warranted to give a comprehensive clear image of the situation in our community.

Implementation of detailed endoscopic grading systems and achalasia subtype classification would enhance diagnostic precision. Measuring patient-reported outcomes using validated score would strengthen assessment methodology.

ACKNOWLEDGEMENTS

We are grateful to the patients and their families for their support of the study.