Every month, JBJS publishes a review of the most pertinent and impactful studies published in the orthopaedic literature during the previous year in 13 subspecialties. Click here for a collection of all such OrthoBuzz summaries.
This month, co-author Nitin B. Jain, MD, MSPH selected the most clinically compelling findings from the 40 studies summarized in the November 20, 2019 “What’s New in Orthopaedic Rehabilitation.”
–A randomized controlled trial compared pain-related function, pain intensity, and adverse effects among 240 patients with chronic back, hip, or knee pain who were randomized to receive opioids or non-opioid medication.1 After 12 months, there were no between-group differences in pain-related function. Statistically, the pain intensity score was significantly lower in the non-opioid group, although the difference is probably not clinically meaningful. Adverse events were significantly more frequent in the opioid group.
–A series of nested case-control studies found that the use of the NSAID diclofenac was associated with an increase in the risk of myocardial infarction in patients with spondyloarthritis and osteoarthritis, relative to those taking the NSAID naproxen.2
–Intra-articular injections of corticosteroids or hyaluronic acid are often used for pain relief prior to an eventual total knee arthroplasty (TKA). An analysis of insurance data found that patients who had either type of injection within three months of a TKA had a higher risk of periprosthetic joint infection (PJI) after the operation than those who had injections >3 months prior to TKA.
Partial-Thickness Rotator Cuff Tears
–A randomized controlled trial of 78 patients with a partial-thickness rotator cuff compared outcomes of those who underwent immediate arthroscopic repair with outcomes among those who delayed operative repair until completing 6 months of nonoperative treatment, which included activity modification, PT, corticosteroid injections, and NSAIDs.3 At 2 and 12 months post-repair, both groups demonstrated improved function relative to initial evaluations. At the final follow-up, there were no significant between-group differences in range of motion, VAS, Constant score, or ASES score. Ten (29.4%) of the patients in the delayed group dropped out of the study due to symptom improvement.
Stem Cell Therapy
–A systematic review that assessed 46 studies investigating stem cell therapy for articular cartilage repair4 found low mean methodology scores, indicating overall poor-quality research. Only 1 of the 46 studies was classified as excellent, prompting the authors to conclude that evidence to support the use of stem cell therapy for cartilage repair is limited by a lack of high-quality studies and heterogeneity in the cell lines studied.
- Krebs EE, Gravely A, Nugent S, Jensen AC, DeRonne B, Goldsmith ES, Kroenke K, Bair MJ, Noorbaloochi S. Effect of opioid vs nonopioid medications on pain-related function in patients with chronic back pain or hip or knee osteoarthritis pain: the SPACE randomized clinical trial. JAMA. 2018 Mar 6;319(9):872-82.
- Dubreuil M, Louie-Gao Q, Peloquin CE, Choi HK, Zhang Y, Neogi T. Risk ofcmyocardial infarction with use of selected non-steroidal anti-inflammatory drugs incpatients with spondyloarthritis and osteoarthritis. Ann Rheum Dis. 2018 Aug;77(8): 1137-42. Epub 2018 Apr 19.
- Kim YS, Lee HJ, Kim JH, Noh DY. When should we repair partial-thickness rotator cuff tears? Outcome comparison between immediate surgical repair versus delayed repair after 6-month period of nonsurgical treatment. Am J Sports Med. 2018 Apr;46(5):1091-6. Epub 2018 Mar 5.
- Park YB, Ha CW, Rhim JH, Lee HJ. Stem cell therapy for articular cartilage repair: review of the entity of cell populations used and the result of the clinical application of each entity. Am J Sports Med. 2018 Aug;46(10):2540-52. Epub 2017 Oct 12.
Every month, JBJS publishes a review of the most pertinent and impactful studies published in the orthopaedic literature during the previous year in 13 subspecialties. Click here for a collection of all such OrthoBuzz summaries. This month, Matthew R. Schmitz, MD, JBJS Deputy Editor for Social Media, selected the most clinically compelling findings from the 50 studies summarized in the October 16, 2019 “What’s New in Shoulder and Elbow Surgery.
Rotator Cuff Repair
–A randomized controlled trial compared immediate and delayed surgical repair of partial-thickness rotator cuff tears.1 No differences in retear rates were found, suggesting that a trial of nonoperative management remains appropriate for partial-thickness tears.
–The search continues for biologic augmentations to improve healing after rotator cuff repair. A study that randomized patients to weekly human growth hormone injections for 3 months or no injections after repair of a large tear found no difference in healing rates.2 Another randomized study of the effect on cuff-repair healing of platelet-rich plasma in a fibrin matrix found no improvement.3 A similar randomized trial of platelet-rich plasma plus thrombin in patients with a single-row repair of the supraspinatus found no differences in clinical outcomes or healing rates.4
–Psychosocial factors have been associated with pain relief and functional improvement after rotator cuff repairs. A longitudinal cohort study found that higher fear-avoidance behavior and alcohol use of ≥1 to 2 times per week compared with alcohol use ≤2 to 3 times per month negatively impacted shoulder pain and function at 18 months postoperatively.5
Osteochondritis Dissecans of the Capitellum
–A study evaluated predictors of success of nonoperatively treating patients with osteochondritis dissecans of the capitellum who did not have fluid underneath the fragment.6 Researchers found that lesion healing was associated with the following:
- Smaller overall lesion size
- No clear margins of the fragment on MRI
- Absence of cyst-like lesions
The authors include a nomogram that clinicians can use to predict healing.
–A study investigated baseball position-specific factors affecting return to play after ulnar collateral ligament (UCL) reconstruction.7 Investigators found the following:
- Position players returned to play sooner than pitchers, but they had lower rates of return to play.
- Catchers had the lowest likelihood of return to play (58.6%) and pitchers had the highest (83.7%).
These findings could help clinicians set expectations for players undergoing UCL reconstruction.
- Kim YS, Lee HJ, Kim JH, Noh DY. When should we repair partial-thickness rotator cuff tears? Outcome comparison between immediate surgical repair versus delayed repair after 6-month period of nonsurgical treatment. Am J Sports Med.2018 Apr;46(5):1091-6. Epub 2018 Mar 5.
- Oh JH, Chung SW, Oh KS, Yoo JC, Jee W, Choi JA, Kim YS, Park JY. Effect of recombinant human growth hormone on rotator cuff healing after arthroscopic repair: preliminary result of a multicenter, prospective, randomized, open-label blinded end point clinical exploratory trial. J Shoulder Elbow Surg.2018 May;27(5):777-85. Epub 2018 Jan 11.
- Walsh MR, Nelson BJ, Braman JP, Yonke B, Obermeier M, Raja A, Reams M. Platelet-rich plasma in fibrin matrix to augment rotator cuff repair: a prospective, single-blinded, randomized study with 2-year follow-up. J Shoulder Elbow Surg.2018 Sep;27(9):1553-63. Epub 2018 Jul 9.
- Malavolta EA, Gracitelli MEC, Assunção JH, Ferreira Neto AA, Bordalo-Rodrigues M, de Camargo OP. Clinical and structural evaluations of rotator cuff repair with and without added platelet-rich plasma at 5-year follow-up: a prospective randomized study. Am J Sports Med.2018 Nov;46(13):3134-41. Epub 2018 Sep 20.
- Jain NB, Ayers GD, Fan R, Kuhn JE, Baumgarten KM, Matzkin E, Higgins LD. Predictors of pain and functional outcomes after operative treatment for rotator cuff tears. J Shoulder Elbow Surg.2018 Aug;27(8):1393-400.
- Niu EL, Tepolt FA, Bae DS, Lebrun DG, Kocher MS. Nonoperative management of stable pediatric osteochondritis dissecans of the capitellum: predictors of treatment success. J Shoulder Elbow Surg.2018 Nov;27(11):2030-7.
- Camp CL, Conte S, D’Angelo J, Fealy SA. Following ulnar collateral ligament reconstruction, professional baseball position players return to play faster than pitchers, but catchers return less frequently. J Shoulder Elbow Surg.2018 Jun;27(6):1078-85. Epub 2018 Mar 23.
There are 15 references to JBJS studies in the recently published 149-page white paper on “Biological Responses to Metal Implants,” from the FDA’s Center for Devices and Radiological Health. Most of those references are made in Section 7.5.1 (pp. 54-57), which focuses on orthopaedic devices.
The plethora of JBJS references is not surprising, but we were happy also to see that a JBJS “Case Connections” article was cited twice in the white paper. While most of the section on orthopaedic devices discussed metal-on-metal (MoM) hip problems, the FDA noted that adverse biological responses to metals in orthopaedics sometimes occur in the upper extremity. It did so by citing “Adverse Local Tissue Reactions in the Upper Extremity,” which appeared in the May 24, 2017 issue of JBJS Case Connector. The FDA white paper cautioned that metal wear debris-related adverse reactions have occurred with shoulder suture anchors (five cases of which are described in the “Case Connections” article) and with intramedullary humeral nailing (one case of which is described in the “Case Connections” article).
Among the take-home points made by co-authors Thomas Bauer and Allan Harper in the cited “Case Connections” article is this: “Patients with shoulder suture anchors who develop delayed-onset pain and/or stiffness, osteolysis, chondrolysis, or early arthropathy should be evaluated and consideration should be given to the removal of loose or prominent anchors to lessen the risk of articular damage.”
We’re all familiar with the phrase “lesser of two evils,” but I’m an optimist and prefer the phrase “better of two goods.” In the October 2, 2019 issue of JBJS, Ramme et al. compare surgical versus nonsurgical treatment of full-thickness rotator cuff tears. Both cohorts had improved outcomes relative to baseline, but surgical management was the better of two goods.
The authors retrospectively analyzed a prospective cohort of adult patients with full-thickness rotator cuff tears who had elected either surgical or nonsurgical treatment. Ramme et al. utilized propensity score matching to pair up patients in each group according to factors thought to influence outcome, such as age, sex, tear size, chronicity, muscle atrophy, and the Functional Comorbidity Index. This matched-pair analysis is a valiant attempt to eliminate bias that is inherent in retrospective analyses, and this study design also mimics the real-world scenario of shared decision making between physician and patient.
The 2-year follow-up analysis of 107 propensity score-matched patients revealed that both groups improved in 4 patient-reported functional outcomes and pain compared to their baseline measures before treatment. However, the final outcome measurements and magnitude of improvement were statistically greater in the surgical management group (p <0.001).
This study will help shoulder surgeons have more meaningful discussions with their patients about treatment options for full-thickness rotator cuff tears. We know that with proper treatment—either surgical or nonsurgical—patients can expect improvement in pain and function. However, patients who elect surgical management may have the potential for even greater outcomes, and that definitely sounds like the “better of two goods.”
Matthew R. Schmitz, MD
JBJS Deputy Editor for Social Media
This post comes from Fred Nelson, MD, an orthopaedic surgeon in the Department of Orthopedics at Henry Ford Hospital and a clinical associate professor at Wayne State Medical School. Some of Dr. Nelson’s tips go out weekly to more than 3,000 members of the Orthopaedic Research Society (ORS), and all are distributed to more than 30 orthopaedic residency programs. Those not sent to the ORS are periodically reposted in OrthoBuzz with the permission of Dr. Nelson.
Advanced glycation end products (AGEs) form through a nonenzymatic process by which reducing sugars undergo Maillard rearrangement with amino acids. During rearrangement, the carbonyl group of the sugar reacts with the amino group of the amino acid, producing N-substituted glycosylamine and water.
During cooking, glycation occurs at 140° to 165° C (280° to 330° F), resulting in the browning of foods such as bread and French fries. This nonenzymatic reaction also occurs at human body temperature over decades. AGE formation can decrease the viscoelasticity and tensile strength of human tissue, resulting in increased mechanical stiffness that affects bone, ligaments, cartilage, and menisci. In cartilage, the excessive accumulation of AGEs leads to a more brittle matrix that is susceptible to fatigue and failure. AGEs also contribute to the etiology of several diabetic complications, including adhesive capsulitis of the shoulder.
Rotator cuff degeneration and tears become more common with age. Accumulated mechanical loads and anatomic variation play a large role. The role of AGEs in rotator cuff degeneration and tears has been suspected, but the exact mechanisms remain in question. Investigators recently showed that AGEs have detrimental effects on human rotator cuff-derived cells in vitro and on intact rat infraspinatus tendons ex vivo.1
In Vitro Findings
Rotator cuff-derived cells were obtained from 12 torn cuff edges during supraspinatus tendon repairs in patients with an average age 64.8 years. The cells were cultured in (1) regular medium with 500 μg/mL AGEs (high-AGE group), (2) regular medium with 100 μg/mL AGEs (low-AGE group), and (3) regular medium alone (control group). Cell viability was significantly suppressed in the high-AGE group relative to the control group. Vascular endothelial growth factor secretion was significantly greater in the high- and low-AGE groups than in the control group. Immunofluorescence stain demonstrated enhancement of hypoxia-inducible factor-1α, reactive oxygen species expressions, and cell apoptosis in the high- and low-AGE groups compared with the control group.
Ex Vivo Findings
Four upper limbs with intact rotator cuff tendons were harvested from 10-week old rats and cultured in regular medium or regular medium with 500 μg/mL AGEs. Mechanical testing showed significantly higher tensile strength in the control group than in the AGE group.
These results beg the question as to whether reduction of AGEs might delay or prevent rotator cuff senescence-related degeneration.
- Mifune Y, Inui A, Muto T, Nishimoto H, Kataoka T, Kurosawa T, Yamaura K, Mukohara S, Niikura T, Kokubu T, Kuroda R. Influence of advanced glycation end products on rotator cuff. J Shoulder Elbow Surg. 2019 Aug;28(8):1490-1496. doi: 10.1016/j.jse.2019.01.022. Epub 2019 Apr 10. PMID: 30981546
It goes almost without saying that a patient’s return to work after an orthopaedic injury or musculoskeletal disorder would correlate with the severity of the condition. But what about the connection between return to work and a more “touchy-feely” parameter, such as the patient-surgeon relationship?
Dubert et al. conducted a longitudinal observational study of 219 patient who were 18 to 65 years of age and had undergone operations for upper-limb injuries or musculoskeletal disorders. In the August 7, 2019 issue of JBJS, they report that a positive relationship between patient and surgeon hastened return to work and reduced total time off from work.
At the time of enrollment (a mean of 149 days after surgery), the authors assessed the patient-surgeon relationship with a validated, 11-item questionnaire called Q-PASREL, and they collected patients’ functional and quality-of-life scores at the same time. The authors then tracked which patients had returned to work 6 months later, and they calculated how many workdays those who did return had missed.
The Q-PASREL questionnaire explores surgeon support provided to the patient, the patience of the surgeon, the surgeon’s appraisal of when the patient can return to work, the cooperation of the surgeon regarding administrative issues, the empathy perceived by the patient, and the surgeon’s use of appropriate vocabulary.
Here is a summary of the findings:
- At 6 months after enrollment, 74% of patients who had returned to work had given their surgeon a high or medium-high Q-PASREL score. By contrast, 64% of the patients who had not returned to work had given their surgeon a low or medium-low Q-PASREL score.
- The odds of returning to work were 56% higher among patients who gave surgeons the highest Q-PASREL scores compared with those who gave surgeons the lowest scores.
- The “body structure” subscore on one of the functional measurements and the Q-PASREL quartile were the only two independent predictors of total time off from work among patients who had returned to work.
After asserting that their study “confirms that surgeons’ relationships with their patients can influence the patients’ satisfaction and outcomes,” Dubert et al. go on to suggest that the findings should prompt surgeons to “work on empathy, time spent with their patients, and communication.” While they rightly claim that such improvements would entail “little financial investment and no side effects,” perhaps the authors, who practice in France, underestimate the effort that goes into changing behavior—and into addressing the time constraints imposed by the US health care system?
It has been estimated that 13% to 16% of patients who undergo arthroscopic stabilization procedures for recurrent shoulder instability are dissatisfied with their outcome, despite a technically “successful” operation. Similarly high rates of patient dissatisfaction in the face of an objectively “well-done” surgery are pervasive in most orthopaedic subspecialties and often leave both surgeon and patient frustrated and perplexed. Prior research has suggested that patient expectations, psychological characteristics, and socioeconomic factors play a major role in these cases of patient dissatisfaction. But identifying precise patient or injury factors that can alert surgeons as to which patients may be unsatisfied after their procedure has remained elusive for many common injuries.
In the June 19, 2019 issue of The Journal, Park et al. examine the bases for patient dissatisfaction after arthroscopic Bankart repair (with or without remplissage) for recurrent shoulder instability. Not surprisingly, patient age, size of the glenoid bone defect, and the number of patient postoperative instability events correlated with an objective failure of the operation (i.e., instability requiring a repeat operation). However, the study found that the number of instability events and the preoperative width of the Hill-Sachs lesion correlated with the subjective failure of the operation (i.e., the patient was dissatisfied based on response to a single question about “overall function” 2 years after surgery). For the 14 out of 180 patients who were dissatisfied despite not experiencing a revision, intermittent pain plus psychological characteristics such as apprehension and anxiety about recurrent instability were common reasons for dissatisfaction.
It is becoming clearer with each passing year that simply correcting anatomic pathologies does not always result in happy patients. Orthopaedic surgeons need to employ patient interviewing techniques to identify issues such as anxiety, depression, pain-perception concerns, and substance abuse—all of which can negatively influence the degree of patient satisfaction with the result and are somewhat modifiable preoperatively.
Marc Swiontkowski, MD
The surgical options for treating irreparable tears of the supraspinatus—cuff reconstruction, tendon transfers, and shoulder replacement—are limited and complicated. But biomechanical results from a cadaveric study of 14 shoulders by Lobao et al., published in the June 5, 2019 issue of JBJS, suggest that a biodegradable balloon spacer inserted subacromially could effectively treat such insufficiencies, possibly postponing the need for more aggressive procedures.
Using an irreparable supraspinatus tear model and sophisticated instruments, the authors determined that, at postoperative time 0, the saline-inflated balloon:
- Restored intact-state glenohumeral contact pressures at most abduction angles
- Moved the humeral head inferiorly by a mean of 6.2 mm at 0° of abduction and 3.0 mm at 60°
- Increased deltoid load by 8.2% at 0° and by 11.1% at 60°.
The balloon, however, did not restore glenohumeral contact area to that of an intact shoulder.
Although the authors cite a previous clinical case series using this approach,1 they are quick to point out that “it is not possible to correlate our findings with clinical scenarios.” Nevertheless, they say that the biomechanical data obtained from this cadaveric study “suggest that the balloon may be of benefit clinically, at least in the immediate postoperative setting.”
- Deranlot J, Herisson O, Nourissat G, Zbili D, Werthel JD, Vigan M, Bruchou F. Arthroscopic subacromial spacer implantation in patients with massive irreparable rotator cuff tears: clinical and radiographic results of 39 retrospectives cases. Arthroscopy. 2017 Sep;33(9):1639-44. Epub 2017 Jun 8
In the setting of rotator cuff injuries, higher degrees of fatty infiltration into cuff muscles are positively correlated with higher repair failure rates and worse clinical outcomes. MRI continues to be the gold standard imaging modality for evaluating fatty infiltration of the rotator cuff, but ultrasound represents another viable modality for that assessment—at considerably lower cost. Such is the conclusion of Tenbrunsel et al. in a recent issue of JBJS Reviews.
The authors reviewed 32 studies that investigated imaging modalities used to assess fatty infiltration and fatty atrophy. They found that grading fatty infiltration using ultrasound correlated well with grading using MRI. However, the authors identified difficulties distinguishing severe from moderate fatty infiltration on ultrasound, but they added that discerning mild from moderate fatty infiltration is more important clinically. Tenbrunsel et al. also mention sonoelastography, which measures tissue elasticity and can also be used to help determine the severity of fatty atrophy of the rotator cuff.
Overall, the trade-off between MRI and ultrasound comes down to higher precision with the former and lower cost with the latter.
For more information about JBJS Reviews, watch this video featuring JBJS Editor-in-Chief Dr. Marc Swiontkowski.
Every month, JBJS publishes a review of the most pertinent and impactful studies published in the orthopaedic literature during the previous year in 13 subspecialties. Click here for a collection of all OrthoBuzz Specialty Update summaries.
This month, Albert Gee, MD, a co-author of the April 17, 2019 “What’s New in Sports Medicine,” selected the five most clinically compelling findings from among the 30 noteworthy studies summarized in the article.
Anterior Cruciate Ligament (ACL) Reconstruction
–Norwegian researchers randomized 120 patients to undergo either single-bundle or double-bundle ACL reconstruction and followed them for 2 years.1 They found no difference between the 2 techniques in any patient-reported outcome, knee laxity measurements, or activity levels. These results, along with the preponderance of evidence from other comparative trials over the last 5 years, strongly suggest that routine use of 2 bundles to primarily reconstruct a torn ACL adds no clinical benefit over a well-positioned single-bundle reconstruction.
Knee Cartilage Repair
–A randomized study compared long-term patient outcomes after knee cartilage repair using microfracture versus mosaicplasty.2 Included patients had 1 or 2 focal femoral lesions measuring between 2 and 6 cm2. Better outcomes after a minimum of 15 years of follow-up were found in the mosaicplasty group. Although there were only 20 patients in each arm, the Lysholm-score differences between the groups were both clinically important and statistically significant. More patients in the mosaicplasty group than in the microfracture group said they would have the surgery again, knowing their 15-year outcome.
–UK researchers randomized 313 patients with ≥3 months of subacromial pain and an intact rotator cuff who had completed a nonoperative program of physical therapy and injection to 1 of 3 groups: arthroscopic subacromial decompression, diagnostic arthroscopy (“sham” surgery), or no intervention.3 At 6 months and 1 year, all groups demonstrated statistically significant and clinically important improvement, but patient-reported outcome scores were significantly better in both surgical groups compared with the no-treatment group. The data suggest that patients such as these improve over time, regardless of management, but that surgical decompression may offer a slight benefit over nonoperative management because of the placebo effect.
–A randomized controlled trial investigated the effect of a formal preoperative education program (2-minute video plus handout)4 about postoperative narcotic use, side effects, dependence risk, and addiction potential among >130 patients undergoing arthroscopic rotator cuff repair surgery. The education group consumed 33% less narcotic medication at 6 weeks and 42% less at 12 weeks compared with the control group. Among the more than one-quarter of the patients who had used opioids prior to surgery, those randomized to the education group were 6.8 times more likely than controls to discontinue narcotic use during the study period.
–A randomized controlled trial of >300 patients compared hip arthroscopy and “best conservative care” for treating femoroacetabular impingement (FAI).5 Only 8% of patients crossed over from conservative care to the surgical group. The mean adjusted difference in iHOT-33 scores at 1 year was 6.8, in favor of hip arthroscopy. However, adverse events were more frequent in the arthroscopy cohort, and a within-trial economic evaluation suggested that hip arthroscopy was not cost-effective compared with conservative care during the 1-year trial period.
- Aga C, Risberg MA, Fagerland MW, Johansen S, Trøan I, Heir S, Engebretsen L. No difference in the KOOS Quality of Life Subscore between anatomic double-bundle and anatomic single-bundle anterior cruciate ligament reconstruction of the knee: a prospective randomized controlled trial with 2 years’ follow-up. Am J Sports Med.2018 Aug;46(10):2341-54. Epub 2018 Jul 18.
- Solheim E, Hegna J, Strand T, Harlem T, Inderhaug E. Randomized study of long-term (15-17 years) outcome after microfracture versus mosaicplasty in knee articular cartilage defects. Am J Sports Med.2018 Mar;46(4):826-31. Epub 2017 Dec 18.
- Beard DJ, Rees JL, Cook JA, Rombach I, Cooper C, Merritt N, Shirkey BA, Donovan JL, Gwilym S, Savulescu J,Moser J, Gray A, Jepson M, Tracey I, Judge A, Wartolowska K, Carr AJ; CSAW Study Group. Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial. Lancet. 2018 Jan 27;391(10118):329-38. Epub 2017 Nov 20.
- Syed UAM, Aleem AW, Wowkanech C, Weekes D, Freedman M, Tjoumakaris F, Abboud JA, Austin LS. Neer Award 2018: the effect of preoperative education on opioid consumption in patients undergoing arthroscopic rotator cuff repair: a prospective, randomized clinical trial. J Shoulder Elbow Surg.2018 Jun;27(6):962-7. Epub 2018 Mar 26.
- Griffin DR, Dickenson EJ, Wall PDH, Achana F, Donovan JL, Griffin J, Hobson R, Hutchinson CE, Jepson M,Parsons NR, Petrou S, Realpe A, Smith J, Foster NE; FASHIoN Study Group. Hip arthroscopy versus best conservative care for the treatment of femoroacetabular impingement syndrome (UK FASHIoN): a multicentre randomised controlled trial. Lancet. 2018 Jun 2;391(10136):2225-35. Epub 2018 Jun 1.