Patient-Specific Pessaries for Pelvic Organ Prolapse Using Three-Dimensional Printing: A Pilot Study

Abstract

Importance

Vaginal pessaries are an effective nonsurgical treatment for pelvic organ prolapse (POP) when properly fitted. However, pessary fitting and use are often unsuccessful or imperfect.

Objective

The objective of this study was to assess the feasibility of using patient-specific pessaries fabricated from three-dimensional (3D)-printed molds to improve POP symptoms and increase overall satisfaction of pessary treatment in patients using standard vaginal pessaries.

Study Design

Patients undergoing POP treatment with standard vaginal pessaries were enrolled in this pilot prospective study. Patient-specific pessaries were designed and fabricated for each patient using patient input, physician input, and anatomic measurements from clinical assessment. Pessary fabrication involved injection of biocompatible liquid silicone rubber into 3D-printed molds followed by a biocompatible silicone coating. Pelvic organ prolapse symptomatic distress and pessary treatment satisfaction were evaluated before and after a 3-week patient-specific pessary home trial using the validated Pelvic Organ Prolapse Distress Inventory-6 form and a visual analog scale, respectively.

Results

Eight women were included in this study. Changing from standard pessary to patient-specific pessary treatment was associated with an improvement in prolapse symptoms on the Pelvic Organ Prolapse Distress Inventory-6 (median change, −3.5; interquartile range, −5 to −2.5; P = 0.02) and an increase in overall pessary satisfaction on a visual analog scale (median change, +2.0; interquartile range, +1.0 to +3.0; P = 0.02). All patients reported either an improvement or no change in pessary ease of use, comfort, and the feeling of support provided by the pessary.

Conclusion

Patient-specific vaginal pessaries are a promising alternative to standard pessaries for alleviating POP symptoms and improving patient satisfaction with pessary use.

WHY THIS MATTERS

Vaginal pessaries are an effective nonsurgical treatment option for patients with symptomatic pelvic organ prolapse, but up to 30% of patients with pelvic organ prolapse who opt for pessary treatment are unable to be successfully fitted with standard pessaries. Personalized, patient-specific pessaries fabricated using 3D printing may be a potential solution for these patients and those imperfectly fitted with standard pessaries. Despite several recent case reports on 3D-printed pelvic floor devices, the field of gynecology is trailing other subspecialty fields in developing and using customized prostheses to improve patient care. This pilot feasibility study helps to bridge that gap by assessing the use of 3D-printed, patient-specific pessaries to improve pelvic organ prolapse symptoms and increase overall satisfaction with pessary use in patients who have been previously fitted with standard pessaries. The results, which show an improvement in prolapse symptoms on the Pelvic Organ Prolapse Distress Inventory 6 questionnaire and overall satisfaction on a visual analog scale after patient-specific pessary use, demonstrate the feasibility of pessary personalization to improve current pessary treatment. By leveraging 3D printing technology together with rapid advancements in artificial intelligence, there is potential for a paradigm shift in the design of vaginal pessaries based on patient-specific anatomy and guided by empiric data rather than trial-and-error.

Simply Stated

Pelvic organ prolapse, the descent of 1 or more pelvic organs (eg, bladder, uterus) into the vagina, is a common and stigmatized medical condition that greatly impairs quality of life for millions of patients worldwide. Vaginal pessaries, prosthetics placed in the vagina, are an effective nonsurgical treatment for pelvic organ prolapse; however, many patients are unsuccessfully or imperfectly fitted with pessaries. Current pessaries are manufactured according to a limited set of shapes and sizes, which may compromise their use. In this pilot study of 8 patients, we explored the feasibility of using 3D-printed pessaries to improve POP symptoms and increase overall satisfaction in patients using standard pessaries. Customized, 3D-printed pessaries were developed for each patient according to their treatment priorities and individual anatomy. A 3-week trial period with the customized pessaries resulted in an improvement in prolapse symptoms and overall satisfaction on standardized questionnaires. All patients preferred treatment with a customized, patient-specific pessary over a standard pessary. This study demonstrates that pessary personalization is feasible and a promising avenue toward improving patient satisfaction and effectiveness with pessary treatment, especially for patients in whom standard pessaries are ineffective or suboptimal in relieving prolapse symptoms.

Vaginal pessaries are an effective and safe nonsurgical treatment option for patients with symptomatic pelvic organ prolapse (POP),¹ a common and stigmatized medical condition that greatly impairs quality of life for millions of patients worldwide.^2–4 Contemporary pessaries are manufactured according to a limited set of shapes and sizes using mass production methods,⁵ which may compromise their efficacy rate. As a result, up to 30% of patients with POP who opt for pessary treatment are unable to be successfully fitted with standard pessaries.⁶ Moreover, among pessary users, many still experience poor fit or undesired sequelae from pessary use (eg, discomfort, ulcerations).⁷ At present, a well-fitting pessary determined by clinicians through a trial-and-error process remains elusive for many patients, thus highlighting an urgent need for pessary treatment personalization.

Three-dimensional (3D) printing is a fabrication method that allows for pessary personalization compared with traditional manufacturing techniques.⁸ Using 3D printing, pessaries’ geometry (eg, shape, size) and mechanical properties (eg, stiffness, bending force) can be designed according to individual patients’ unique anatomy and goals of pessary care. A personalized, patient-specific pessary may therefore be a potential nonsurgical treatment option for patients who are unsuccessfully or imperfectly fitted with standard devices.⁹

The objective of this study was to assess the feasibility of using Gynethotics^TM (patient-specific pessaries) to improve POP symptoms and to increase overall satisfaction with pessary use in patients who have been previously fitted with standard pessaries.

MATERIALS AND METHODS

This study was reviewed and approved by the institutional research ethics board (20-0248-A). An investigational testing authorization for the use of the patient-specific pessaries in this study was obtained from Health Canada (321095). Written informed consent was obtained from all participants before enrollment in the study.

Patient Recruitment

Patients with symptomatic POP attending office-based care at a single academic-based urogynecology practice unit were recruited via letters between October and December 2021. Patients were eligible for inclusion if they had been successfully fitted with a standard pessary to manage POP and if they had used and self-managed the pessary for at least 6 months before study enrollment. Exclusion criteria were as follows: active vaginal erosions, ulcerations, lacerations, bleeding, posterior-predominant prolapse (Pelvic Organ Prolapse Quantification [POP-Q] point Bp > point Ba or C), symptoms of urinary or genital tract infection, neurologic disease associated with loss of pelvic floor sensation, inability to ambulate, inability to verbally communicate with examiner, pregnancy, or younger than 18 years.

Data Collection

Patient demographics, standard pessary characteristics, and patient-reported symptoms related to pessary use were collected at the initial visit by a urogynecologist. This included assessment of prolapse symptoms using the Pelvic Organ Prolapse Distress Inventory-6 (POPDI-6)¹⁰ and overall satisfaction level with patients’ current pessary using a visual analog scale (VAS) (Supplemental materials, Appendix A, http://links.lww.com/FPMRS/A395).¹¹ Patients were asked to identify the top 3 priority features in a pessary and to provide recommendations for changes to their current pessary that would improve their experience with pessary usage (Supplemental materials, Appendix B, http://links.lww.com/FPMRS/A395). Patients then underwent a baseline pelvic examination, which included assessment of prolapse using the POP-Q system, vaginal capacity by lateral and anterior-posterior finger spread, and levator ani muscle strength by digital assessment (modified Oxford scale). As part of a concurrent pilot study investigating the use of a novel imaging technique to assess vaginal capacity, all patients underwent 3D transintroital ultrasonography with a polyurethane vaginal bag in situ to capture vaginal shape, volume, and pressure during intravaginal distension with water.

Patient-Specific Pessary Customization

Pessary Design

Three patient-specific pessaries were developed for each patient based on patient input, physician input, and anatomic measurements from clinical assessment. For example, if a patient reported suboptimal comfort with her current ring pessary but was unable to retain the next smallest size pessary, a patient-specific pessary shape and size would be developed based on anatomical measurements and physician input. Similarly, if a patient reported ease of insertion and removal to be a high priority feature, factors that facilitated ease of use (ie, lower durometer silicone and more flexible bending region) would be prioritized in patient-specific pessary design. The physician performing the baseline examination noted current pessary characteristics that, if modified, could potentially help achieve the patient's desired improvements with pessary usage. In addition, transintroital ultrasound volumes of the distended vagina were used to establish maximum thresholds for pessary dimensions.

Patient-Specific Pessary Production

To fabricate the patient-specific pessaries, the geometry of each pessary was first developed using a commercial 3D computer-aided design software (Fusion 360 by Autodesk, CA). Next, a negative cocoon mold of the design was 3D printed in 1 part using a 3D printer optimized for biocompatible materials (Formlabs, MA). Compared with the generally used clamshell molds,⁹ cocoon molds consume less material and thus generate less waste. The resin used for the mold was BioMed Clear (material number: FLBMCL01), which was chosen for its biocompatibility. Then, medical-grade liquid silicone rubber (LSR, Elkem SILBIONE, East Brunswick, NJ) was injected into the mold. Once the LSR was cured, it was removed from the mold and the pessary was postcured at 200 ± 10°C for 4 hours. Next, a medical-grade silicone coating (NuSil Technology, Carpinteria, CA) was applied (Fig. 1). All devices were carefully inspected according to manufacturing quality specifications, cleaned with 70% isopropanol, and packaged before use in this study. The silicone used in patient-specific pessary fabrication has a lower durometer rating (ie, softer and more flexible) compared with silicone in standard pessaries.

FIGURE 1.

Sequential steps in patient-specific pessary fabrication. (A), A three-dimensional (3D) computer-aided (CAD) design model of a ring with support (used as an example here) is created. Labels A–F in the CAD design correspond to pessary dimensions that can be independently customized. (B), A cocoon mold is 3D printed based on the CAD model. (C), The mold is injected with medical-grade liquid silicone rubber and cured. (D), The mold is removed, revealing the patient-specific pessary.

Patient-Specific Pessary Treatment and Follow-Up

After fabrication of the patient-specific pessaries, participants presented for a fitting to ensure an adequate fit of 3 patient-specific designs. Successful fitting was defined as intravaginal retention of 1 or more of the patient-specific pessaries with reasonable comfort during a Valsalva maneuver. Participants then wore each patient-specific pessary for 1 week, during which they kept a diary of pessary care and comfort qualities, as well as pelvic floor symptoms associated with patient-specific pessary use relative to their standard pessary. Patients were also instructed to monitor for pessary-related adverse events, including vaginal bleeding, abnormal discharge or odor, pain, or irritation. The home trial period was 3 weeks in total. After this period, patients returned to the clinic and identified the most favorable of the 3 patient-specific pessaries; this was used for outcomes comparison with their standard pessary. Patients then completed a follow-up questionnaire, which included assessment of pelvic floor symptoms, overall satisfaction, and qualitative feedback on the patient-specific pessary (Supplemental materials, Appendix C, http://links.lww.com/FPMRS/A395). A pelvic examination was also performed to evaluate for pessary-related adverse events. To receive further patient feedback, all patients were provided an opportunity for a second round of patient-specific pessary fitting if desired.

The primary outcome was the change in the POPDI-6 score from the initial assessment (with standard pessary use) to follow-up assessment (after patient-specific pessary use). The POPDI-6 is scored on a scale from 0 to 24, with higher scores representing worsening prolapse distress. Secondary outcomes included the following: changes in the overall satisfaction VAS score, patient assessment of pessary qualities and pelvic floor symptoms, patient preference for a standard or patient-specific pessary, and qualitative patient feedback on pessary design and function.

Data Analysis

Demographic characteristics and patient responses to categorical items on questionnaires were described using descriptive statistics, as appropriate. Data were expressed as median with interquartile range (IQR). Changes in the POPDI-6 score and differences in the overall satisfaction VAS score between standard and patient-specific pessaries were assessed using the Wilcoxon sign-rank test. A P value of 0.05 was considered statistically significant. All data were analyzed using STATA (v17.0; College Station, TX).

RESULTS

Demographics and Clinical Measurements

Eight women with a median age of 62 years (range, 42–72.5 years) were included in the study. Demographic and clinical information collected before patient-specific pessary treatment are shown in Table 1. All participants had POP-Q stage 2 or greater prolapse. At the initial assessment, the most common type of standard pessary used was the ring with support, followed by the Marland, the cube, and the Gellhorn (Table 2). The median duration of pessary use before study enrollment was 36 months (IQR, 23–73 months). At baseline (ie, initial assessment with standard pessary use), the median POPDI-6 score was 7.5 (IQR, 4.5–8.5) and the median VAS score was 6.0 (IQR, 5.0–8.5).

TABLE 1.

Demographic and Clinical Characteristics of Study Patients

Variables	Patients (N = 8)
Age, y	62.0 (42.0–72.5)
BMI	24.0 (20.5–25.5)
Vaginal parity	2.0 (2.0–2.0)
POP-Q stage
Stage 2	6 (75%)
Stage 3	1 (12%)
Stage 4	1 (12%)
Vaginal estrogen use
No	3 (38%)
Yes	5 (62%)
Occult urinary incontinence from pessary use
No	5 (62%)
Yes	3 (38%)
POPDI-6	7.5 (4.5–8.5)
VAS	6.0 (5.0–8.5)

BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); POP-Q, Pelvic Organ Prolapse Quantification; POPDI-6, Pelvic Organ Prolapse Distress Inventory-6; VAS, visual analog scale.

Data are expressed as median with IQR or number of patients (% of total patients) unless otherwise specified.

TABLE 2.

Characteristics of Standard Pessaries Used by Patients Before Study Enrollment and Patient-Specific Pessaries Used in the Study

Patient	Standard Pessary		Patient-Specific Pessary
	Type	Dimensions (Length, Width, Height)	Type	Dimensions (Length, Width, Height)	Example of Patient-Desired Pessary Features	Example of Patient-Centered Design Decisions
P1	Circular ring with support	83, 83, 12	Oval ring with support	91, 80, 15	Desired a stiffer, more supportive pessary for exercise	Ring thickness increased from 12 mm to 15 mm
P2	Circular ring with support	70, 70, 12	Oval ring with support	77, 68, 12	Desired a softer pessary	Softer silicone used for fabrication (durometer 40 rating)
P3	Circular ring with support and knob	70, 70, 12	Oval ring with support and knob	75, 70, 13	Desired improved anti-incontinence	Pessary length increased to position knob under the mid-urethra
P4	Circular ring with support and knob	76, 76, 12	Oval ring with support and knob	86, 76, 12	Desired a pessary that is easier to fold	Pessary notch along the bending axis increased by 40%
P5	Marland	89, 89, 44	Marland	89, 89, 39	Desired a smaller pessary for improved comfort	Elevated platform height decreased from 44 mm to 39 mm
P6	Marland	64, 64, 37	Marland	74, 68, 37	Desired improved fit with reduced dislodgement	Pessary dimensions customized to optimize fitting
P7	Cube	44, 44, 44	Oval ring with support	74, 68, 14	Desired to improve self-maintenance (previously failed ring placement)	Oval ring pessary dimensions customized to optimize fitting
P8	Gellhorn	57, 57, 8.5	Gellhorn	62, 52, 9	Desired to improve self-maintenance; desired less irritation to posterior vagina from Gellhorn stem	Pessary dimensions customized to facilitate self-maintenance; stem angled 20 degrees anteriorly

For ring Marland, and Gellhorn pessaries, length is represented by the maximum dimension of the disc portion. Height is represented by the vertical dimension of the pessary normal to the disc (ie, ring thickness, Marland horn height, or Gellhorn stem length). Units are in millimeters. The patient-specific pessary presented is the one identified as most favorable by the patient after the trial period. Changes in outcomes for individual patients are depicted in Figure 2. Three patients (P1, P2, and P8) elected to undergo a second round of patient-specific pessary treatment.

POP Distress and Overall Satisfaction Outcomes After Patient-Specific Pessary Treatments

All 8 patients were successfully fitted with a patient-specific pessary. Individual changes in POPDI-6 and VAS scores from baseline to status post patient-specific pessary trial are shown in Figure 2. Averaged values are summarized in Table 3. The POPDI-6 score decreased (ie, reflected improvement in prolapse distress) in 7 of the 8 patients after treatment with a patient-specific pessary (median change, −3.5; IQR, −5 to −2.5; P = 0.02), whereas overall satisfaction VAS score increased for 7 of the patients (median change, +2.0; IQR, +1.0 to +3.0; P = 0.02). One patient reported maximum pessary use satisfaction (VAS score = 10) for both standard and patient-specific pessaries. Three patients elected to undergo a second round of patient-specific pessary treatment to provide further feedback on customized pessary designs.

FIGURE 2.

Individual changes in POP distress (POPDI-6) and overall satisfaction (VAS) scores with standard pessary use (baseline) and after patient-specific pessary use (posttreatment). POPDI-6: ranging from no distress (0) to high distress (24), VAS, ranging from low satisfaction (0) to high satisfaction (10). Specific pessary characteristics for each patient are described in Table 2. POPDI-6, Pelvic Organ Prolapse Distress Inventory-6; VAS, visual analog scale.

TABLE 3.

Changes in Distress and Overall Satisfaction Scores After Patient-Specific Pessary Use

Variables	Patients (N = 8)
POPDI-6	−3.5 (−5.0 to −2.5)*
VAS	2.0 (1.0–3.0)*
Patient-reported outcomes
Ease of care
Better	4 (50%)
Same	4 (50%)
Worse	0
Comfort
Better	4 (50%)
Same	4 (50%)
Worse	0
Feeling of support
Better	4 (50%)
Same	4 (50%)
Worse	0

POPDI-6, ranging from no distress (0) to high distress (24); VAS, ranging from low satisfaction (0) to high satisfaction (10).

Data are expressed as median with IQR or number of patients (% of total patients) unless otherwise specified. * P < 0.05 in Wilcoxon sign-rank test.

POPDI-6, Pelvic Organ Prolapse Distress Inventory-6; VAS, visual analog scale.

All patients reported either improvement or no change in pessary ease of use, comfort, feeling of support provided by the pessary, and prolapse symptoms (Table 3). One patient reported dislodgement of the patient-specific pessary during a bowel movement. None of the patients reported pessary-related adverse events, and no vaginal erosions, bleeding, or abnormal discharge was observed during the follow-up speculum examination at the clinic.

In response to our request for comments on their patient-specific pessaries, participants provided comments that were largely favorable toward a patient-specific design:

“The current pessary felt like I was wearing a tampon and I felt pressure. The [patient-specific pessary] was not felt in me most of the time.” “Better fit so it stays in position” and “Loved how flexible the material was, made removal and insertion much easier.” Patients also provided feedback for improvements: “I would prefer that it gave the feeling of more support of prolapse.” and “Old Gellhorn was hard to insert and remove, but more effective.”When asked “Would you rather use [a patient-specific pessary] or your pessary?” on the questionnaire, all 8 patients indicated preference for a patient-specific pessary.

DISCUSSION

In this pilot feasibility study of 8 women using standard pessaries for symptomatic POP, we found that switching to a patient-specific pessary was associated with a decrease in prolapse symptoms, as assessed by the POPDI-6 form, and an increase in overall satisfaction with pessary use, as assessed by a VAS scale. Our findings suggest that patient-matched pessaries, generated from patient-centered 3D-printed designs, have the potential to improve prolapse treatment outcomes compared with standard pessaries.

With rapid advancement and increasing accessibility of 3D printing technology within the past decade, recent studies have demonstrated proof of principle for customized, 3D-printed vaginal prosthesis to treat pelvic floor conditions in the absence of other suitable treatments. In 2018, Barsky et al⁹ reported the fabrication of a customized incontinence ring with knob, using a 3D-printed mold, for a 90-year-old nulligravid patient with stress incontinence in whom commercial anti-incontinence pessary fitting had failed. Their customized pessary was found to be comfortable by the patient and effective in eliminating urinary leakage during a 48-hour trial period. More recently, Kisby et al¹² demonstrated the feasibility of using a 3D-printed intravaginal mold to maintain vaginal patency after McIndoe neovagina creation in a patient with a unique pelvic shape in whom handheld vaginal dilator use had failed. Nevertheless, the field of gynecology is trailing other subspecialty fields, such as orthodontics¹³ or audiology¹⁴ in developing and using customized prostheses to improve patient care.

Our current work expands upon prior research in 3D printing for gynecologic applications in several novel ways. First, this study demonstrates the feasibility of using pessaries made through 3D printing for treatment of POP, which has not previously been shown. Although the study period was brief, no pessary-related adverse events were observed during the study. Second, in contrast to prior feasibility studies, our study involved patients who had been successfully fitted with a standard pessary prior to enrollment in the study. Our results highlight an opportunity to improve the current approach to pessary design and manufacturing. In our study, even the patient who reported a maximum VAS score of 10 for overall satisfaction with her standard pessary reported improvement in prolapse symptoms and ease-of-use with a patient-specific pessary design.

Third, we used a patient-centered design approach, which incorporated specific patient feedback in the design process for a patient-specific pessary. Ideally, all features of a pessary (eg, reduction of prolapse symptoms, ease of self-care or comfort) would be perfectly realized. In reality, each design decision is associated with tradeoffs.¹⁵ By including patient-specific feedback in the design process, patient-specific pessaries can be engineered to optimize all features in alignment with patients' treatment priorities. For example, our study included a patient using a Gellhorn pessary who highly desired to improve self-maintenence, so a patient-specific pessary that facilitated self-insertion and self-removal was designed (Table 2, P8). Although the patient reported dislodgement of the patient-specific pessary during a bowel movement, she favored the patient-specific pessary because it served her treatment priorities. Similarly, a patient using a standard cube pessary (Table 2, P7) favored a patient-specific ring with support pessary, despite a 1-point increase in her POPDI-6 score, because the patient-specific pessary allowed for improved self-maintenence.

Fourth, although prior studies have reported subjective symptom improvement and physical examination observations after treatment with customized pelvic devices, this study incorporated use of the validated POPDI-6 form for assessing changes in prolapse symptoms. Although changes in POPDI-6 scores were modest in this cohort of current pessary users, we expect the reduction in prolapse distress to be more pronounced among patients who cannot be fitted or are unhappy with standard pessaries. Findings from this study will be helpful to inform effect size and minimally clinical important difference calculations in larger prospective studies involving patient-specific pessaries for patients in whom standard pessary use has failed.

Finally, this study involved a cohort of 8 patients, which is the largest cohort undergoing a trial of personalized, 3D printed pelvic devices to date. Although this remains a small cohort, it demonstrates the feasibility of a personalized design approach beyond individual patients.

Currently, pessary selection and fitting are based on a “trial-and-error” process and feedback loop, which is subjective and dependent on a clinician’s experience. By leveraging 3D-printing technology, as demonstrated in this study, together with rapid advancements in artificial intelligence, ¹⁶ there is potential for a paradigm shift in the design of vaginal pessaries based on patient-specific anatomy and guided by empiric data. The application of machine learning-based techniques may help determine, for example, the optimal shape, dimensions, and mechanical properties of a pessary that will maximize treatment success, as defined by the patient and health care provider input.

This study should be interpreted in the context of several important limitations. First, patients were not blinded. Future studies incorporating randomization and/or blinding processes may be useful to determine changes in treatment efficacy and patient satisfaction without influences of the placebo effect from using a “personalized” product. Nevertheless, our finding that all 8 patients reported preference for the patient-specific pessary over their standard pessary is self-evident. Second, we did not include patients in whom prior pessary placement had failed. We expect that a patient-specific design approach would improve treatment success by reducing discomfort, incomplete resolution of symptoms, or difficulty with pessary self-management. In these patients, a successfully fitted patient-specific pessary may be a value-based alternative to surgical treatment. However, we recognize that some patients may never be successfully fitted with a pessary, no matter the design. In addition, although patient-specific pessaries were specifically designed for individual patients, they remain modifications of existing design concepts (eg, ring, Gellhorn). With ongoing research in pessary and pelvic floor biomechanics and continued advancements in 3D printing technologies, we anticipate the ability to develop new classes of adaptive pessaries^17,18 for an expanding global population of pessary users (“Pessary Market Size & Share, Industry Report, 2021–2028,” GVR-4-68038-299-0).¹⁹ These pessaries may have applications beyond the treatment of pelvic organ prolapse, including use in obstetrics, the management of stress urinary incontinence, and as prophylaxis for the pelvic floor during weight-bearing activities.

We have demonstrated that personalization of pessary design for the treatment of POP is feasible and a promising avenue toward improving patient satisfaction and clinical efficacy of pessary treatment, especially for those in whom standard pessaries are ineffective or suboptimal in relieving pelvic floor symptoms. In an era of personalized medicine, further utilization of 3D printing within obstetrics and gynecology will likely improve patient care, as it has in other fields.

ACKNOWLEDGMENTS

The authors would like to thank Drs. Michael Chaikof, Maria Giroux, and Humara Edell for their help with the clinical examination, Ravnit Lomash and Etsuko Tsuchiya for their assistance with administrative work and patient recruitment, and Maria Lacalle-Aurioles for her assistance with manuscript preparation.