Disclosure: The author has no financial relationship with any orthopaedic device manufacturer.
================================================================
There is increasing interest in the use of hemiarthroplasty — rather than total shoulder arthroplasty (TSA) — in the management of cuff-intact arthritis. The ream-and-run (RnR) is a glenohumeral arthroplasty in which a cobalt-chromium humeral hemiarthroplasty is combined with conservative reaming of the glenoid bone to a single concentric concavity. The reamed bone remodels during the rehabilitation period to provide a biological bearing surface that articulates directly with the prosthetic head, eliminating the modes of failure unique to a prosthetic glenoid — wear, loosening, and bone loss. [1]
The radiograph above shows wear and loosening of a glenoid component with surrounding bone loss.
The RnR can enable high levels of shoulder function without the risk of plastic glenoid component failure.
It allows biological remodeling of challenging arthritic deformities such as the bad B2 shown below.
The RnR requires no proprietary planning software and no special components or instruments. It can be performed with most commercially available shoulder arthroplasty systems. The outcome depends on careful patient selection, attention to the details of surgical technique, and a dedication to the rehabilitation program. [1,2]
This is a review of what we think we know about the ream and run.
Sections:
1. The Beginning
2. Indications and Patient Selection
3. Surgical Technique
4. Time Course of Recovery
5. Glenoid Wear and Radiographic Outcomes
6. Comparison with Anatomic Total Shoulder Arthroplasty
7. Complications and Reoperation
8. Cost and Value
9. Predictors of Success
10. Not for Every Patient, Every Surgeon, or Every Problem
11. Will the Outcomes for Pyrocarbon Hemiarthroplasty Be Different from Those for the RnR?
12. Conclusions
Additional information on the Ream and Run origins, patient and surgeon, surgery, rehabilitation, and results can be found at the ream and run website.
================================================================
1. THE BEGINNING
Recognition of the potential value of non-prosthetic glenoid arthroplasty emerged from observing that patients with failed total shoulder replacements recovered comfort and function after the worn or loosened polyethylene glenoid component was removed, leaving behind a smooth bony concavity.
In a canine model, the reamed glenoid bone became covered with conforming, securely attached fibrocartilaginous tissue at 24 weeks, with progressive subsurface trabecular bone densification consistent with physiologic load transfer. [20]
Slides above show (a) normal glenoid, (b) reamed glenoid, (c) surface regeneration at 12 wks, and (d) surface regeneration at 24 wks.
Lynch and colleagues reported the first prospective RnR clinical series showing significant improvement in self-assessed comfort and function (mean SST 4.7 ± 2.4 preoperatively to 9.4 ± 2.6 at 2–4 year follow-up), with no surgical complications, no infections, no instability, and no revisions to total shoulder arthroplasty. [18] Another analysis compared RnRs with TSAs by the same surgeon, matched for diagnosis, sex, age, side, and follow-up duration. TSA reached its functional plateau 12 to 18 months earlier, but from 18 through 36 months mean SST scores did not differ significantly between procedures (36-month SST 9.5 vs 10.0). [21] This finding was confirmed by subsequent comparative studies. [8]
================================================================
2. INDICATIONS AND PATIENT SELECTION
RnR is an alternative to aTSA for patients with cuff-intact primary glenohumeral osteoarthritis, capsulorrhaphy arthropathy, or post-traumatic arthritis who wish to avoid the long-term limitations associated with a polyethylene glenoid component. [2] The prototypical RnR candidate is described as a motivated, resilient male patient with glenoid biconcavity and retroversion who wants to participate in heavy lifting and impact sports. [16] In the UW 544-case concurrent series, patients who chose RnR were more often male (92.0% vs 47.0%, p < .001), younger (mean 58 ± 9 vs 67 ± 10 years, p < .001), more likely to be married (83.2% vs 66.8%), more frequently from out of state (51.7% vs 21.7%), and significantly less likely to be on narcotic pain medication preoperatively (16.6% vs 26.6%, p = 0.005). [6] Their shoulders also differed: RnR patients had substantially higher prevalence of Walch B2 glenoids (46.0% vs 27.8%, p < .001), greater preoperative retroversion (19 ± 11° vs 15 ± 11°, p < .001), and higher preoperative SF-36 physical functioning scores (73 ± 17 vs 56 ± 23, p < .001) than aTSA patients. [6] In practice, RnR is the procedure offered to (and chosen by) the more biomechanically challenging shoulders in the more physically active patients.
Severe glenoid pathoanatomy does not preclude good outcomes. In a 49-shoulder cohort from an external high-volume center, 30 of 49 (61%) had Walch type B glenoids and patient-reported outcomes were not associated with Walch classification or with preoperative shoulder motion. [4] A matched-cohort analysis comparing RnR with aTSA likewise found no significant differences in outcome attributable to glenoid morphology after concentric reaming. [8]
Figure . RnR populations are enriched for B2 (biconcave) glenoids across institutions. (A) Distribution of Walch subtypes in 263 RnR and 281 aTSA patients at the University of Washington. [6] B2 glenoids — the morphology most often considered challenging for arthroplasty — are present in 46% of RnR patients but only 28% of aTSA patients at the same institution. (B) B2 prevalence converges across independent RnR cohorts (UW 46.0% [6]; Virginia Beach 41.0%, combined from 47 self-selecting and 31 control patients [19]), both well above the UW aTSA comparator (27.8%, dotted line). At every reporting center, RnR populations carry the more biomechanically challenging glenoid morphology.
================================================================
3. SURGICAL TECHNIQUE
The shoulder is approached through an anterior deltopectoral incision.
The subscapularis is peeled from the lesser tuberosity, retaining the capsule on the deep surface of the tendon.
A 360-degree release of the subscapularis restores the excursion of the muscle.
The humeral head is completely resected, avoiding an inadequate resection as shown on the left and protecting the rotator cuff as shown on the right. The plane of resection is 30 degrees of retroversion and 45 degrees with the long axis of the humeral shaft.
The glenoid is reamed with a nubbed (rather than cannulated) reamer. Reaming is conservative, only enough to establish a single concavity — prioritizing preservation of the glenoid bone stock. No attempt is made to alter glenoid version.
With the trial humeral component in place, the range of motion is verified, using the 150-40-50-60 guidelines: ≥150° of flexion, ≥40° external rotation with the subscapularis approximated to its reattachment site, ≥50% manual humeral translation, and 60° internal rotation in abduction. [1,2]
The subscapularis is securely repaired to the lesser tuberosity. Assisted forward flexion is started immediately after surgery to preserve the 150° of flexion achieved at surgery.
We inform the patient before surgery that if they have difficulty maintaining their range of motion, a manipulation under anesthesia with complete muscle relaxation can be considered.
================================================================
4. TIME COURSE OF RECOVERY
The recovery for each of the 12 Simple Shoulder Test functions at two years after surgery is shown below.
The trajectory of recovery after RnR is one of the most clinically useful pieces of preoperative counseling information. Gilmer and colleagues, analyzing 176 consecutive RnR procedures with longitudinal follow-up, demonstrated that shoulder comfort and function climb rapidly during the first year, continue to improve more slowly through the second year, and reach a steady state at approximately 20 months. [3] Among patients with at least 2 years of follow-up, 124 of 140 (89%) achieved the minimal clinically important difference (MCID) on the Simple Shoulder Test (SST). [3] The plateau is durable: Stenson and colleagues reported a mean SST of 10.2 at 5 years [5], and Sharareh and colleagues reported a mean SST of 10.3 with 82% of patients achieving MCID at 10 years [7].
Figure 2. Recovery trajectory and durability of patient-reported improvement after ream-and-run. Early time points (0–24 months) from Gilmer [3]; 5-year point from Stenson [5]; 10-year point from Sharareh [7]. Dotted vertical line marks the 20-month plateau.
================================================================
5. GLENOID WEAR AND RADIOGRAPHIC OUTCOMES
A historical objection to any humeral hemiarthroplasty has been that the prosthetic humeral head will progressively erode the glenoid, producing late pain and dysfunction. The modern quantitative wear literature does not support that objection.
Somerson and colleagues first characterized the clinical and radiographic course of RnR for primary glenohumeral osteoarthritis: medialization of the humeral head center of rotation occurs predominantly early and stabilizes, and radiographic medialization does not correlate with patient-reported outcomes. [9] Collins and colleagues extended this work in 113 RnR shoulders followed for a mean of 6.7 ± 2.3 years (range 4.2–11.4 years), measuring medialization on standardized AP radiographs. [10] The data were best fit by a quadratic function (R² = 0.82, p = .001) rather than a linear one: wear averaged 0.6 mm/year during the first 4 years and decelerated to 0.2 mm/year between years 4 and 10, reaching a cohort mean of 2.9 ± 4.3 mm at 6.7 years (Figure 3A). Wear was minimal or mild (≤5 mm) in 81% of patients, moderate (>5 to ≤10 mm) in 13%, and severe (>10 mm) in only 5% (Figure 3B).
The finding that matters clinically: wear severity did not predict patient-reported outcomes. SST, change in SST, VAS, and change in VAS were statistically indistinguishable between minimal/mild and moderate/severe wear cohorts (all p > 0.17). [10] Open revision rates were higher in the moderate/severe group (23.8% vs 7.6%, p = .004), but the absolute revision rate for the full cohort remained low at 10.6% over a mean of 6.7 years.
Figure 3. Glenoid wear after ream-and-run decelerates after year 4 and does not track with patient-reported outcomes (Collins 2026, n = 113, mean 6.7 yr follow-up). (A) Glenoid medialization over time. Curve = quadratic fit to longitudinal data (R² = 0.82, p = .001); dot = observed cohort mean at 6.7 years (2.9 mm); dotted reference line = expected trajectory if wear were linear at 0.3 mm/yr. Mean wear rate was 0.6 mm/yr during years 0–4 and 0.2 mm/yr during years 4–10. (B) Severity distribution at most recent follow-up: 81% minimal/mild (≤5 mm), 13% moderate (>5 to ≤10 mm), 5% severe (>10 mm); totals reflect rounding. Patient-reported outcomes (SST, ΔSST, VAS, ΔVAS) did not differ between mild and moderate/severe wear cohorts (all p > 0.17).
================================================================
6. COMPARISON WITH ANATOMIC TOTAL SHOULDER ARTHROPLASTY
Levins and colleagues conducted a propensity-matched cohort study comparing humeral-head replacement with concentric glenoid reaming (RnR) against aTSA in 39 matched pairs of younger patients (all male, <66 years of age, mean 58.6 ± 7.3 years; >75% with B- or C-type glenoids) at mean follow-up of 4.4 ± 2.3 years. [8] On paired t-tests, aTSA had statistically better final SST (10.9 vs 10.3) and ASES (89.9 vs 85.0) scores than RnR, but in the mixed-effects model that controlled for baseline covariates, arthroplasty type was not associated with any difference in PROM, HRQoL, or %MPI outcome. At 2 years, more aTSA patients achieved the MCID for VAS pain (89.7% vs 75%) and the SCB for ASES (100% vs 79.2%); however, at ≥5-year follow-up, MCID, SCB, and PASS achievement were statistically indistinguishable between cohorts on all three measures (SST, ASES, VAS pain). The Sharareh minimum 10-year analysis provides the parallel long-term picture: at the time horizon at which aTSA glenoid loosening typically becomes clinically relevant (9–14 years), RnR delivered SST and pain outcomes equivalent to aTSA without exposure to the prosthetic glenoid as a late failure mode. [7]
A JBJS commentary noted that the Levins analysis represents the largest comparative study of RnR outside the originating institution and the most rigorously matched comparison in the literature. [16] This commentary identifies limitations of the Levins propensity-score matching and argues that, on balance, these limitations strengthen rather than weaken the case for RnR equivalence. First, the matching algorithm could not capture activity level or psychosocial attributes such as resilience and motivation — characteristics that, if anything, favor the RnR cohort. If RnR patients are systematically more resilient (as Levins and colleagues subsequently demonstrated [15]) and the matched analysis still shows equivalence, the procedure is performing well before that resilience advantage is accounted for. Second, matching on duration of follow-up loaded the aTSA arm with more patients from the later (post-2011) period, when RnR was an option — so the aTSA cohort may include patients who actively chose aTSA over an available RnR alternative. Both limitations argue that the observed equivalence is a conservative estimate. [16]
The two procedures carry distinctly different risk profiles in time. In the Levins matched cohort, three RnR patients underwent revision for pain at a mean of 1.9 ± 1.7 years (two with positive Cutibacterium cultures at revision), while two aTSA patients required revision for glenoid loosening at 9.2 and 14 years. [8] RnR carries a higher early revision rate (largely for pain or stiffness within the first 2 years); aTSA defers risk to later glenoid-component failure. [7,8] A 544-case concurrent series likewise documented that, despite the marked differences in patient and shoulder characteristics, the two procedures produced clinically similar outcomes: mean 2-year SST of 10.0 ± 2.6 (RnR) vs 9.5 ± 2.7 (aTSA), with percent of maximum possible improvement of 72 ± 39% vs 73 ± 29%, respectively. [6]
A counterintuitive finding from the Levins inverse-probability-weighted sensitivity analysis of all 167 patients deserves note: dissatisfaction was significantly higher after aTSA (22.9%) than after RnR (9.4%), despite the higher RnR early revision rate. [8] The authors attribute this to either (1) the absence of postoperative activity restrictions for RnR patients, (2) deterioration in aTSA outcomes at longer follow-up in younger patients, (3) selection bias whereby patients with later aTSA problems are more likely to return for evaluation, or (4) that patients self-selecting RnR may also be more likely to rate themselves "satisfied" because of treatment-preference effects. The finding is only hypothesis-generating but it is consistent with the long-term equivalence findings.
Suttmiller, Snyder, and Carofino contributed independent confirmation from a single-surgeon practice outside the UW system: in a Virginia Beach cohort of 46 shoulders (23 RnR and 23 aTSA, all male, mean age 56.2 ± 8.3 years) matched 1:1, there were no differences in PROMs between RnR and aTSA at 1- and 2-year follow-up, although RnR patients reported significantly higher daily pain ratings (p = 0.047) and lower ASES scores (p = 0.031) at 3 months — consistent with the well-documented slower early recovery of RnR. [11]
Mostafa and colleagues subsequently performed a PRISMA-compliant systematic review and meta-analysis of comparative RnR versus aTSA studies (738 RnR / 810 aTSA across 8 studies). The pooled analysis found no significant differences between RnR and aTSA in SST, ASES, VAS, or forward flexion. A higher rate of return to theatre was reported for RnR (7.0%), consistent with the early-revision profile noted in the UW data. [12] An independent 2025 PRISMA meta-analysis by Roelker and colleagues, pooling 668 shoulders across three comparative studies, reached the same conclusion: significant pre-to-postoperative improvements in SST, ASES, VAS pain, and range of motion in both arms, with no statistically significant difference between RnR and aTSA in any outcome measure. The RnR revision rate (11.7%) was more than two and a half times that of aTSA (4.4%) when only short-term outcomes were considered. [17]
Figure 4. Convergent evidence for clinical equivalence between ream-and-run and anatomic total shoulder arthroplasty. Forest plot of mean SST difference (RnR − aTSA, 95% CI) across five independent comparative studies — a propensity-matched cohort [8], a single-institution long-term cohort [7], a non-UW matched comparison [11], and two PRISMA meta-analyses [12,17] totaling more than 2,200 patients across multiple study designs and institutions. None of the five comparisons reaches statistical significance; all confidence intervals cross zero. Both meta-analyses report higher short-term revision rates for RnR (7.0–11.7%) than aTSA (4.4%).
================================================================
7. COMPLICATIONS AND REOPERATION
Stiffness is the most common postoperative complication after RnR and the leading indication for early reoperation. Schiffman and colleagues analyzed 340 RnR patients and found a combined intervention rate of 17.9% (7.6% open revision plus 10.3% manipulation under anesthesia at mean follow-up of 2.1 years). On multivariate analysis, younger age (OR 0.96/year, p = 0.040), ASA class 1 compared with class 3 (OR 0.14, p = 0.020), and less passive forward elevation at the time of hospital discharge (OR 0.96/degree, p < 0.001) were independent predictors of reoperation for stiffness. [13] A clinically important secondary finding was that 69.2% (18 of 26) of patients undergoing open revision for stiffness had at least 2 positive intraoperative cultures for Cutibacterium — indicating that occult low-grade infection may be associated with stiffness and is one possible contributor among several to the observation that young, healthy patients appear more prone to reoperation for stiffness. [13] These data inform both patient selection and preoperative counseling: patients with these risk factors should be told explicitly that their probability of a second procedure for stiffness is higher than average, and surgeons should maintain a low threshold for sending intraoperative cultures at the time of any revision for stiffness. [13]
The complication landscape on the aTSA side looks very different, both in timing and in what failure looks like. aTSA carries failure modes distributed across the temporal spectrum rather than deferred to a single late peak. Of aTSA reoperations attributable to instability, rotator cuff failure, and infection, approximately 63% occur within the first 2 years. [23] After 2 years, aTSA fails at a steady linear rate of approximately 1.1% per year for all causes combined, with mechanical failure (aseptic loosening, component wear, and implant fracture) accumulating progressively and accounting for 41% of all aTSA reoperations (85 of 208), and rotator cuff failure accounting for an additional 22% (45 of 208). [23]
In the young patient population most directly relevant to the RnR comparison, the cumulative revision burden of aTSA is substantially higher than the often-cited general-population figures — Neyton's 202-patient cohort reaches 40% cumulative revision at 20 years, with glenoid failure accounting for 88% of revisions and a precipitous decline in survivorship after 10 years. [22] Severe glenoid wear after RnR, by contrast, is uncommon (5% in the Collins cohort [10]), and when it does occur it is not by itself an indication for revision in the absence of clinical symptoms. [9,10] Other historical young-patient series have reported similar long-term revision rates. [26–30]
Figure 5. Cumulative revision rate after primary anatomic total shoulder arthroplasty over 20 years. The red curve anchors the figure: Neyton 2019 [22] reports Kaplan-Meier survivorship of 95% / 83% / 60% at 5 / 10 / 20 years in 202 patients aged ≤60 undergoing primary aTSA for primary glenohumeral osteoarthritis, equivalent to cumulative revision rates of 5%, 17%, and 40%, with glenoid loosening accounting for 88% of revisions. The three gray curves are long-term comparators provided for context: Schoch 2015 [28] (minimum 20-year follow-up in patients aged <50, mixed diagnoses) and Evans 2021 [29] (minimum 20-year follow-up of the Aequalis prosthesis in a mixed-age cohort) each report cumulative revision rates of approximately 16% at 20 years; the Evans 2020 systematic review and meta-analysis [31] of case-series and national registry data pools mixed-age patients and reports approximately 8% at 10 years. The central observation is that the long-term revision burden of aTSA in young patients with primary osteoarthritis (Neyton 40% at 20 years) substantially exceeds the figures commonly cited from mixed-age long-term series (~16% at 20 years) and from registry pools (~8% at 10 years) — a difference of clinical importance when counseling younger active patients about the comparative durability of aTSA versus ream-and-run.
Figure 6. Revision risk after RnR and aTSA in younger patients: timing and consequence are not the same.
(A) Cumulative revision incidence over 20 years. The aTSA curve is anchored to Neyton 2019 [22], a multicenter Kaplan-Meier analysis of 202 patients aged ≤60 undergoing primary aTSA for primary glenohumeral osteoarthritis: 95% / 83% / 60% revision-free at 5 / 10 / 20 years, equivalent to cumulative revision rates of 5%, 17%, and 40%, with glenoid loosening accounting for 88% of revisions. The 40% at 20 years is at the upper end of published estimates; smaller long-term series in younger aTSA cohorts report revision rates of ≈16% at 20 years (Schoch 2015 [28]; Evans 2021 [29]), although Neyton's is the largest analysis and uses the most rigorous Kaplan-Meier methodology. The RnR open-revision curve is anchored to Schiffman 2023 [13] (7.6% at mean 2.1 years, n=340) and Sharareh 2024 [7] (12% at minimum 10 years, n=34). The aTSA curve crosses the RnR open-revision curve at approximately year 8. Manipulation under anesthesia is the dominant non-open intervention after RnR — Schiffman reports a combined intervention rate of 17.9% at 2.1 years (7.6% open plus 10.3% MUA), and Sharareh reports 14.7% combined at 10 years — but MUA involves no incision, no implant change, and no commitment to a different procedure, and is therefore not plotted on the cumulative revision curve. In the broader aTSA population, reoperations for instability, rotator cuff failure, and infection are concentrated in the first 2 years (63% of reoperations for these three causes combined occur within 2 years). [23,24]
(B) The asymmetry of revision. Equal cumulative revision rates between RnR and aTSA are not equal revision events. When an aTSA is revised, approximately 85% are converted to a reverse total shoulder arthroplasty [25]; the conversion is one-way, since the resected humeral head, the reamed glenoid, and the resurfaced subscapularis preclude re-creation of the original anatomic procedure. When an RnR is revised, the intervention is heterogeneous and largely conservative: in the Schiffman 2023 cohort [13], 35 of 61 intervention events (57%) were MUA only and 26 (43%) were open revisions, typically humeral-head exchange or debridement. Within the open-revision category, most revisions preserve the original procedure rather than escalate to a different one — Sharareh 2024 [7] reported four open revisions among 34 RnR patients at minimum 10-year follow-up: two head exchanges, two head downsizings, and zero conversions to aTSA or rTSA. An RnR revision typically leaves the patient with the original procedure intact and every downstream conversion option still available; an aTSA revision typically commits the patient to a fundamentally different procedure for life.
Figure 7. Indications for open revision among 508 RnRs at the University of Washington, plotted by date of index surgery. Each dot represents one open-revision event. Stiffness is the dominant indication throughout the series, particularly before 2017. After 2017, the overall frequency of open revision decreases and the indication mix shifts. The contributors to this temporal change are not known, but likely include refinements in patient selection, surgical technique, infection-prevention protocols, and rehabilitation.
================================================================
8. COST AND VALUE
Chawla and colleagues examined drivers of inpatient hospital cost and improvement in health-related quality of life across 222 aTSAs and 211 RnRs at a single institution. The RnR procedure was associated with lower inpatient cost — a finding that is straightforward: the RnR uses a single cobalt-chromium humeral head with no glenoid component, no cement, and no patient-specific planning software, so the implant-cost line item is materially smaller than in aTSA. Greater improvements in health-related quality of life were associated with lower preoperative SF-6D and EQ-5D scores and with higher preoperative optimism scores, in both procedures. [14]
Figure 8. Decomposition of inpatient hospital cost across three procedures: ream-and-run (RnR), anatomic total shoulder arthroplasty with standard polyethylene glenoid (aTSA standard), and anatomic TSA with augmented or patient-specific component (aTSA with augmented/PSI). Blue blocks represent cost categories shared across all three procedures and are scaled to typical proportions of total inpatient cost. Red blocks represent aTSA-specific categories absent in RnR — the polyethylene glenoid (added in all aTSA cases) and the augmented or patient-specific component (added in the subset that uses it). The structural cost differential between RnR and aTSA standard is approximately 20%; between RnR and aTSA with augmented/PSI, approximately 32%. Anchored to Chawla 2021 [14]; category proportions are illustrative and consistent with published shoulder arthroplasty cost-decomposition data.
================================================================
9. PREDICTORS OF SUCCESS
Across the literature, several patient factors are associated with favorable outcomes after RnR. Male sex is the most consistent predictor across all follow-up durations and PROMs. Lower preoperative SST scores identify patients with greater potential for measurable gain. Primary osteoarthritis as the diagnosis predicts better outcomes than post-traumatic or capsulorrhaphy arthropathy, and absence of prior shoulder surgery is associated with greater improvement and lower reoperation risk (p < 0.04). [3,5]
Psychosocial factors also matter. Levins, Dasari, and colleagues evaluated patient resilience and mental health as predictors of outcome and found that Connor-Davidson Resilience Scale scores and preoperative VR-12 Mental Component Scores were independently correlated with satisfaction (p < 0.05). [15] RnR patients demonstrated higher mean resilience scores than aTSA patients (34.3 ± 4.8 vs 32.5 ± 6.2, p < 0.001), suggesting that the procedure may attract — or require — more psychologically resilient individuals to navigate the long recovery. [15]
A common concern in the RnR literature is that published results are biased by patients self-selecting the procedure — travelling long distances, researching the technique independently, and pre-committing to a specific surgeon, all of which could plausibly inflate outcomes through a treatment-preference effect. Suttmiller, Snyder, and Carofino tested this hypothesis directly in 78 shoulders from their Virginia Beach practice, comparing the 47 shoulders in patients who deliberately sought out the surgeon and procedure (RnR_SS) against 31 shoulders in patients to whom the surgeon offered RnR after a standard arthroplasty consultation (RnR_CON). [19] Both groups achieved substantial improvements far exceeding the MCID threshold (SST MOI 84–89%, ASES MOI 78%, daily-pain MOI 76–80%) and 66 of 78 shoulders (85%) achieved MCID on all four outcome measures. There were no differences between self-selecting and control patients on any maximum-outcome-improvement metric, any MCID-achievement proportion, or subjective satisfaction (87.2% vs 80.6% "much better", p = 0.569).
Patients who are younger and those who have limited passive forward elevation at the time of hospital discharge are at higher risk of stiffness requiring reoperation and should be counseled accordingly, with attention to aggressive early-postoperative range-of-motion rehabilitation. [13]
Factors that do not predict outcomes include patient age, preoperative shoulder motion, and Walch glenoid classification (type A vs B). [3,4] Severe glenoid retroversion, biconcavity, and posterior subluxation should not, in themselves, be considered contraindications. [4,6]
Figure 9. Predictors of outcome after ream-and-run, synthesized across the cited literature. Positive predictors (top, blue) identify patient and shoulder characteristics associated with favorable outcomes on patient-reported measures and on revision-free survival, organized into demographic, diagnostic, and psychosocial categories. Neutral factors (middle, gray) have been tested in the published literature and found not to predict outcome — notable because several are commonly assumed to do so, including Walch glenoid type A vs B and patient self-selection, the most frequently cited objection to the RnR literature. Negative predictors (bottom, red) identify patient profiles in which RnR carries an elevated risk of stiffness reoperation specifically, not of inferior patient-reported outcome; these patients should be offered the procedure with explicit counseling about that risk and with attention to aggressive early-postoperative range-of-motion rehabilitation. Bar lengths are visual anchors only and do not represent effect sizes; citations indicate the source publications for each predictor.
================================================================
10. NOT FOR EVERY PATIENT, EVERY SURGEON, OR EVERY PROBLEM
A 2015 essay set out the limits of the RnR. RnR requires a patient willing to invest in a 20-month recovery and to commit to a daily home rehabilitation program; a surgeon experienced in conservative concentric reaming and soft-tissue balancing; and a clinical situation in which the trade-off — accepting a higher early revision risk in exchange for elimination of late prosthetic-glenoid complications — is clinically meaningful. [2]
================================================================
11. WILL THE OUTCOMES FOR PYROCARBON HEMIARTHROPLASTY BE DIFFERENT FROM THOSE FOR THE RnR?
As I emphasized in a prior blog post, answering this question will require a prospective trial that controls not only for important patient and shoulder characteristics, but also for the way the glenoid is managed (no glenoid work, selective burring, non-corrective reaming, corrective reaming), as well as humeral head size and position.
Here is my "back of the envelope" estimate. The MCID for the ASES score is between 6 and 14 points and the MCID for the SST is approximately 2 points. The standard deviation at 2 years is approximately 20 points for ASES and 3 points for SST. Assuming alpha = 0.05 (two-sided), 80% power, and 20% loss to follow-up at 2 years, between 100 and 400 matched pairs would be required to demonstrate superiority of either RnR or pyrocarbon hemiarthroplasty over the other, depending on which PROM and which MCID anchor is used. If patients are matched 1:1 on age (±5 years), sex, Walch type, and glenoid management, the cumulative match yield is roughly 25%, meaning initial enrollment would need to be approximately 4× the analyzable sample size — between 400 and 1,600 patients.
A non-inferiority design would be more tractable but would establish only that the implants perform similarly on PROMs — it would not address the comparative value (patient benefit per dollar) of the two procedures, which requires parallel cost analysis.
================================================================
12. CONCLUSIONS
The accumulated evidence supports the ream-and-run as a procedure that delivers durable patient-reported improvement for active patients with cuff-intact glenohumeral arthritis who wish to avoid the long-term risk of prosthetic glenoid failure. The original 2007 prospective series of Lynch and colleagues established the equivalence of self-assessed outcomes to anatomic total shoulder arthroplasty performed by the same surgeon at the same institution. [18] The 20-month recovery trajectory described by Gilmer [3] is reliably followed by a plateau that is sustained through 5 years (Stenson [5]) and 10 years (Sharareh [7]), with outcomes equivalent to aTSA in matched comparisons (Levins [8]), concurrent series [6], an independent external-center cohort (Suttmiller/Carofino [11]), and two pooled meta-analyses (Mostafa [12]; Roelker [17]). Modern quantitative wear data show modest, biphasic, and clinically silent medialization (Collins [10]). RnR is associated with lower inpatient costs (Chawla [14]) and patient resilience is a meaningful predictor of satisfaction (Levins JG [15]). The decision to perform a RnR should be grounded in the procedure's appropriate use — careful patient selection, dedicated technique, and committed rehabilitation — recognizing that it is not for every patient, every surgeon, or every problem. [2]
Getting a grasp
Follow on twitter/X: https://x.com/RickMatsen
Follow on facebook: https://www.facebook.com/shoulder.arthritis
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/
================================================================
REFERENCES
1. Matsen FA III, Carofino BC, Green A, Hasan SS, Hsu JE, Lazarus MD, McElvany MD, Moskal MJ, Parsons IM IV, Saltzman MD, Warme WJ. Shoulder hemiarthroplasty with nonprosthetic glenoid arthroplasty: the ream-and-run procedure. JBJS Reviews. 2021;9(8):e20.00243. doi:10.2106/JBJS.RVW.20.00243. PMID: 34432729.
2. Matsen FA III. The ream and run: not for every patient, every surgeon or every problem. International Orthopaedics. 2015;39(2):255–261. doi:10.1007/s00264-014-2641-2. PMID: 25616729.
3. Gilmer BB, Comstock BA, Jette JL, Warme WJ, Jackins SE, Matsen FA III. The prognosis for improvement in comfort and function after the ream-and-run arthroplasty for glenohumeral arthritis: an analysis of 176 consecutive cases. J Bone Joint Surg Am. 2012;94(14):e102. doi:10.2106/JBJS.K.00486. PMID: 22810409.
4. Levins J, Passarelli E, Adkins J, Molino J, Henry H, Paxton ES, Green A. Early outcome of humeral head replacement with glenoid reaming arthroplasty (ream and run) for treatment of advanced glenohumeral osteoarthritis. J Shoulder Elbow Surg. 2022;31(9):1846–1858. doi:10.1016/j.jse.2022.01.152. PMID: 35276348.
5. Stenson JF, Collins AP, Yao JJ, Sharareh B, Whitson AJ, Matsen FA III, Hsu JE. Factors associated with success of ream-and-run arthroplasty at a minimum of 5 years. J Shoulder Elbow Surg. 2023;32(6S):S85–S91. doi:10.1016/j.jse.2023.01.024. PMID: 36813226.
6. Matsen FA III, Whitson A, Jackins SE, Neradilek MB, Warme WJ, Hsu JE. Ream and run and total shoulder: patient and shoulder characteristics in five hundred forty-four concurrent cases. International Orthopaedics. 2019;43(9):2105–2115. doi:10.1007/s00264-019-04352-8. PMID: 31240359.
7. Sharareh B, Whitson AJ, Matsen FA III, Hsu JE. Minimum 10-year follow-up of anatomic total shoulder arthroplasty and ream-and-run arthroplasty for primary glenohumeral osteoarthritis. J Shoulder Elbow Surg. 2024;33(6):1276–1284. doi:10.1016/j.jse.2023.08.028. PMID: 37777045.
8. Levins J, Molla V, Adkins J, Molino J, Pasarelli E, Paxton ES, Green A. Comparison of humeral-head replacement with glenoid-reaming arthroplasty (ream and run) versus anatomic total shoulder arthroplasty: a matched-cohort study. J Bone Joint Surg Am. 2023;105(7):509–517. doi:10.2106/JBJS.22.00650. PMID: 36727972.
9. Somerson JS, Neradilek MB, Service BC, Hsu JE, Russ SM, Matsen FA III. Clinical and radiographic outcomes of the ream-and-run procedure for primary glenohumeral arthritis. J Bone Joint Surg Am. 2017;99(15):1291–1304. doi:10.2106/JBJS.16.01201. PMID: 28763415.
10. Collins AP, Sheth MM, Stenson JF, Kahsai EA, Khoo KJ, Ogunleye T, Whitson AJ, Matsen FA III, Hsu JE. Characterizing glenoid wear after hemiarthroplasty with concentric glenoid reaming: a study of 113 arthroplasties at a mean of 6.7 years of follow-up. J Shoulder Elbow Surg. 2026;35(5):995–1002. Epub 2025 Oct 8. PMID: 41072717.
11. Suttmiller AMB, Snyder BA, Carofino BC. Comparison of short- and midterm outcomes in patients following ream-and-run and anatomic total shoulder arthroplasties. J Shoulder Elbow Surg. 2025;34(3):794–802. doi:10.1016/j.jse.2024.06.011. PMID: 39103085.
12. Mostafa OES, Jordan RW, Thangarajah T, MacLean S, Woodmass J, D'Alessandro P, Malik SS. Ream-and-run technique offers equivalent clinical outcomes as anatomical total shoulder arthroplasty but with a high rate of complications: a systematic review and meta-analysis. J Orthop. 2025;63:206–215. doi:10.1016/j.jor.2025.04.003. PMID: 40303354.
13. Schiffman CJ, Jurgensmeier K, Yao JJ, Wu JC, Whitson AJ, Jackins SE, Matsen FA III, Hsu JE. Risk factors for stiffness requiring intervention after ream-and-run arthroplasty. JB JS Open Access. 2023;8(2):e22.00104. doi:10.2106/JBJS.OA.22.00104. PMID: 37123506.
14. Chawla SS, Whitson AJ, Schiffman CJ, Matsen FA III, Hsu JE. Drivers of lower inpatient hospital costs and greater improvements in health-related quality of life for patients undergoing total shoulder and ream-and-run arthroplasty. J Shoulder Elbow Surg. 2021;30(8):e503–e516. doi:10.1016/j.jse.2020.10.030. PMID: 33271324.
15. Levins JG, Dasari SP, Quinlan NJ, Whitson AJ, Matsen FA III, Hsu JE. Anatomic shoulder arthroplasty: the correlation between patient resilience, mental health, and outcome. J Shoulder Elbow Surg. 2024;33(6S):S9–S15. doi:10.1016/j.jse.2024.03.008. PMID: 38548096.
16. Hsu JE. Can we reliably compare outcomes of ream-and-run and anatomic total shoulder arthroplasty? Commentary on an article by James Levins, MD, et al.: "Comparison of humeral-head replacement with glenoid-reaming arthroplasty (ream and run) versus anatomic total shoulder arthroplasty. A matched-cohort study." J Bone Joint Surg Am. 2023;105(7):e21. doi:10.2106/JBJS.23.00034. PMID: 37017618.
17. Roelker L, Ghasemi A, Fabregas A, Shafer G, Raphael J. Ream and run hemiarthroplasty versus total shoulder arthroplasty: a comparison of shoulder treatments for glenohumeral arthritis. Cureus. 2025;17(7):e88813. doi:10.7759/cureus.88813. PMID: 40861556.
18. Lynch JR, Franta AK, Montgomery WH Jr, Lenters TR, Mounce D, Matsen FA III. Self-assessed outcome at two to four years after shoulder hemiarthroplasty with concentric glenoid reaming. J Bone Joint Surg Am. 2007;89(6):1284–1292. doi:10.2106/JBJS.E.00942. PMID: 17545432.
19. Suttmiller AMB, Snyder BA, Carofino BC. Patient self-selection does not influence postoperative improvements in pain, function, or satisfaction in ream-and-run arthroplasty patients. J Shoulder Elbow Surg. 2026;35(4):989–994. doi:10.1016/j.jse.2025.08.006. PMID: 40868275.
20. Matsen FA III, Clark JM, Titelman RM, Gibbs KM, Boorman RS, Deffenbaugh D, Korvick DL, Norman AG, Ott SM, Parsons IM IV, Sidles JA. Healing of reamed glenoid bone articulating with a metal humeral hemiarthroplasty: a canine model. J Orthop Res. 2005;23(1):18–26. doi:10.1016/j.orthres.2004.06.019. PMID: 15607870.
21. Clinton J, Franta AK, Lenters TR, Mounce D, Matsen FA III. Nonprosthetic glenoid arthroplasty with humeral hemiarthroplasty and total shoulder arthroplasty yield similar self-assessed outcomes in the management of comparable patients with glenohumeral arthritis. J Shoulder Elbow Surg. 2007;16(5):534–538. doi:10.1016/j.jse.2006.11.003. PMID: 17509900.
22. Neyton L, Kirsch JM, Collotte P, Collin P, Gossing L, Chelli M, Walch G. Mid- to long-term follow-up of shoulder arthroplasty for primary glenohumeral osteoarthritis in patients aged 60 or under. J Shoulder Elbow Surg. 2019;28(9):1666–1673. doi:10.1016/j.jse.2019.03.006. PMID: 31202630.
23. Schoch B, Werthel JD, Schleck CD, Harmsen WS, Sperling J, Sánchez-Sotelo J, Cofield RH. Optimizing follow-up after anatomic total shoulder arthroplasty. J Shoulder Elbow Surg. 2017;26(6):997–1002. doi:10.1016/j.jse.2016.10.024. PMID: 28109672.
24. Werner BC, Burrus MT, Begho I, Gwathmey FW, Brockmeier SF. Early revision within 1 year after shoulder arthroplasty: patient factors and etiology. J Shoulder Elbow Surg. 2015;24(12):e323–e330. doi:10.1016/j.jse.2015.05.035. PMID: 26163282.
25. Gauci MO, Cavalier M, Gonzalez JF, Holzer N, Baring T, Walch G, Boileau P. Revision of failed shoulder arthroplasty: epidemiology, etiology, and surgical options. J Shoulder Elbow Surg. 2020;29(3):541–549. doi:10.1016/j.jse.2019.07.034. PMID: 31594726.
26. Denard PJ, Raiss P, Sowa B, Walch G. Mid- to long-term follow-up of total shoulder arthroplasty using a keeled glenoid in young adults with primary glenohumeral arthritis. J Shoulder Elbow Surg. 2013;22(7):894–900. doi:10.1016/j.jse.2012.09.016. PMID: 23312293.
27. Walch G, Moraga C, Young A, Castellanos-Rosas J. Results of anatomic nonconstrained prosthesis in primary osteoarthritis with biconcave glenoid. J Shoulder Elbow Surg. 2012;21(11):1526–1533. doi:10.1016/j.jse.2011.11.030. PMID: 22445158.
28. Schoch B, Schleck C, Cofield RH, Sperling JW. Shoulder arthroplasty in patients younger than 50 years: minimum 20-year follow-up. J Shoulder Elbow Surg. 2015;24(5):705–710. doi:10.1016/j.jse.2014.07.016. PMID: 25306496.
29. Evans JP, Batten T, Bird J, Thomas WJ, Kitson JB, Smith CD. Survival of the Aequalis total shoulder replacement at a minimum 20-year follow-up: a clinical and radiographic study. J Shoulder Elbow Surg. 2021;30(10):2355–2360. doi:10.1016/j.jse.2021.01.038. PMID: 33675966.
30. Deshmukh AV, Koris M, Zurakowski D, Thornhill TS. Total shoulder arthroplasty: long-term survivorship, functional outcome, and quality of life. J Shoulder Elbow Surg. 2005;14(5):471–479. doi:10.1016/j.jse.2005.02.009. PMID: 16194737.
31. Evans JP, Evans JT, Craig RS, Mohammad HR, Sayers A, Blom AW, Whitehouse MR, Rees JL. How long does a shoulder replacement last? A systematic review and meta-analysis of case-series and national registry reports with more than 10 years of follow-up. Lancet Rheumatol. 2020;2(9):e539–e548. doi:10.1016/S2665-9913(20)30226-5.
