CLINICAL RESEARCH
Custom Reverse Shoulder Arthroplasty for
Severe Postoperative Glenoid Bone Defects: A Retrospective Case Series
Diego J. Gómez,
Nadia Gabotto Loredo, Gonzalo M. Viollaz, Gustavo
Teruya, Álvaro Muratore, Alejandro Tedeschi, Lucio Gutiérrez, Rafael Durán, Santiago Ávila
Posada, Daniel Moya
Upper Limb Team, Hospital Británico de Buenos Aires, Autonomous City of Buenos Aires,
Argentina
ABSTRACT
Introduction: Glenoid
bone loss complicates revision reverse shoulder arthroplasty. Patient-specific
3D-printed glenoid implants allow accurate reconstruction of severe defects. Objective: To
evaluate the clinical and radiographic outcomes of customized glenoid
baseplates in revision reverse shoulder arthroplasty with severe glenoid
defects, and to record intraoperative and post-operative complications. Materials and
Methods: A retrospective series of eight consecutive patients (June
2022–May 2023) with Gohlke type ≥3 defects who
underwent revision reverse shoulder arthroplasty with 3D-planned titanium
glenoid baseplates based on computed tomography. Range of motion, function,
pain, surgical time, and radiographic integration were analyzed, with a
12-month follow-up. Results: Range of motion improved
significantly: forward elevation increased by 78.8° (p < 0.001), external
rotation by 7.5° (p = 0.019), and internal rotation improved from the gluteal
level (trochanter–L5) to a mean of T12 (range: T7–gluteal level). Pain
decreased by 6.4 points (p < 0.001). The Constant-Murley
score increased by 48.9 points (p < 0.001). Complete osseointegration
was observed in 7 cases; the remaining case showed radiolucent lines without
loosening. No major complications were recorded. Conclusions: In revision reverse
shoulder arthroplasty with severe glenoid bone defects, 3D-printed glenoid
base-plates restore anatomy, significantly improve function, and reduce pain at
one year, with a high rate of integration and low morbidity.
Keywords: Reverse
shoulder arthroplasty; glenoid bone defects; patient-specific implants; 3D
printing; revision surgery.
Level of Evidence: IV
Artroplastia inversa de hombro
personalizada para defectos óseos glenoideos severos posoperatorios. Estudio retrospectivo
de casos clínicos
RESUMEN
Introducción: La pérdida ósea glenoidea complica las revisiones de una
artroplastia inversa de hombro. Los implantes glenoideos personalizados
impresos en 3D permiten reconstruir, con precisión, defectos severos. Objetivos: Evaluar
los resultados clínicos y radiográficos de las metaglenas
personalizadas en revisiones de artroplastias inversas de hombro con defectos
glenoideos severos, así como registrar las complicaciones intra
y posoperatorias. Materiales y Métodos: Serie retrospectiva de 8
pacientes consecutivos (junio 2022-mayo 2023) con defectos tipo Gohlke ≥3 sometidos a una artroplastia inversa de hombro de
revisión con metaglenas de titanio planificadas en 3D
sobre una tomografía. Se analizaron la movilidad y la función, el dolor, el
tiempo quirúrgico y la integración radiográfica; seguimiento de 12 meses. Resultados: Los
rangos de movilidad se incrementaron: la elevación anterior aumentó 78,8° (p
<0,001); la rotación externa, 7,5° (p = 0,019); la rotación interna, desde
el nivel glúteo (trocánter-L5) a un promedio de T12 (rango T7-nivel glúteo). El
dolor disminuyó 6,4 puntos (p <0,001). La escala de Constant-Murley
aumentó 48,9 puntos (p <0,001). Se constató la integración ósea completa en
7 casos; el restante tenía líneas radiolúcidas sin
aflojamiento. No se registraron complicaciones mayores. Conclusiones: En las revisiones
de artroplastias inversas de hombro con defectos glenoideos severos, las metaglenas impresas en 3D restauran la anatomía, mejoran
significativamente la función y reducen el dolor al primer año, con una alta
tasa de integración y baja morbilidad.
Palabras clave: Artroplastia inversa de hombro; defectos óseos
glenoideos; implantes personalizados; impresión 3D; revisión quirúrgica.
Nivel
de Evidencia: IV
INTRODUCTION
In
recent decades, reverse shoulder arthroplasty has emerged as an effective
treatment option for rotator cuff tear arthropathy.
Due to its favorable outcomes, its indications have expanded to include
four-part proximal humeral fractures in elderly patients, massive irreparable
rotator cuff tears without arthropathy, and revision
of failed osteosynthesis and hemiarthroplasties.1-3
Loss
of glenoid bone stock is a common and challenging problem when performing
reverse shoulder arthroplasty.4,5 Glenoid bone defects are more frequently found in
revision procedures, due to prosthetic loosening (septic or aseptic), periprosthetic glenoid fractures, and traumatic defects
secondary to migration or protrusion of implants used for proximal humeral
fractures.6-8
Various
complex techniques have been described to address glenoid bone loss, including oneor two-stage revision procedures using iliac crest or
humeral head bone grafts, the use of metal-augmented baseplates, Mark Frankle’s alternative centerline technique, and even
salvage hemiarthroplasty.4,5 More recently, patient-specific implants
manufactured using 3D printing based on precise preoperative digital planning
have been introduced, which adapt to each patient’s specific anatomical defects.
This technology provides a better management of complex glenoid bone defects by
optimizing surgical accuracy and primary implant stability.5–11
Given
the novelty of this technology and the limited available evidence, we have not
yet determined which technique provides the best outcomes.
The
aim of this study was to evaluate the clinical and radiological outcomes at one
year of follow-up in patients undergoing revision reverse shoulder arthroplasty
for severe postoperative glenoid bone defects using patient-specific glenoid
components, as well as to record intraoperative and postoperative
complications.
MATERIALS AND METHODS
A
retrospective analysis was conducted of a consecutive series of patients with
severe glenoid bone defects associated with prior surgery, treated with reverse
shoulder arthroplasty using patient-specific glenoid implants, between June
2022 and May 2023. Adults with severe postoperative glenoid bone defects (≥3
according to the Gohlke classification) and a minimum
clinical and radiographic follow-up of one year were included. Patients
undergoing primary arthroplasty, those with a history of shoulder infection,
and those with neurological deficits in the affected limb were excluded.
Glenoid
defects were classified according to the Gohlke
classification.12
- Type 1: mild bone loss, central or eccentric, with
retroversion <15°.
- Type 2: moderate contained bone loss, with an intact glenoid
vault.
- Type 3: severe eccentric defect with retroversion >20° or
significant loss of glenoid width.
- Type 4: moderate medialization of the glenoid surface.
- Type 5: defect with residual depth <10 mm for implant
fixation.
All
patients underwent a preoperative protocol that included true anteroposterior
and axial radiographs of the affected shoulder, thin-slice computed tomography
(CT) with 3D reconstruction, and laboratory tests including erythrocyte
sedimentation rate and C-reactive protein.
In the
postoperative period, clinical and radiographic evaluations were performed
immediately after surgery, at 1 month, 6 months, 1 year, and at the final
follow-up. Operative time (in minutes) was recorded based on surgical reports,
and intraoperative and postoperative complications were documented during
scheduled follow-up visits.
The
treating surgeon assessed shoulder range of motion preoperatively and at
follow-up visits. Evaluation parameters included active motion, measured as
forward elevation in the scapular plane, external rotation with the elbow at
the side, and internal rotation estimated according to the highest vertebral
level reached by the thumb. In addition, pain was assessed using the visual
analog scale, and function was evaluated using the Constant–Murley
score.
Statistical Analysis
Categorical
variables are expressed as frequencies and percentages, and continuous
variables as mean and standard deviation or median and interquartile range,
depending on their distribution. Normality of differences between preand postoperative values was assessed using the
Shapiro–Wilk test. When normality was confirmed, the paired Student’s t-test
was used; otherwise, the nonparametric Wilcoxon signed-rank test was applied. A
p value
<0.05
was considered statistically significant. All analyses were performed using GraphPad Prism 9.0 (La Jolla, CA, USA).
Custom Implant Design
Preoperative
computed tomography (CT) scans were used for implant design and manufacturing.
Preoperative planning was performed using 3D Slicer (version 5.6.2) and Meshmixer (version 3.5.474, Autodesk Inc., San Rafael, CA,
USA). This allowed detailed evaluation of the bone defect and creation of a 3D biomodel of the patient’s scapula (Figure
1).
The
patient-specific glenoid baseplate was designed in
collaboration with a biomedical engineer, developing a component precisely
adapted to the patient’s bone defect, with the aim of reproducing the glenoid
lateralization, inclination, and version specified by the surgeon (Figure 2).
When
the bone defect was so severe that premorbid lateralization could
not be reliably estimated, superimposition was performed using a CT scan
obtained prior to the initial surgery (if available) or, alternatively, from
the contralateral shoulder (Figure 3).
The
direction and length of the screws were subsequently planned
according to each patient’s bone quality and bone stock. The goal, whenever
possible, was to achieve at least 1 cm of contact between the central peg and
native bone, along with the placement of at least two screws measuring 4.5 mm
in diameter and 30 mm in length (Figure 4).
The
custom baseplate was designed to be Morse taper-compatible with a 36-mm Unique®
glenosphere (Bioprotece,
Villa Ballester, Buenos
Aires, Argentina).
Once
the implant was designed, patient-specific surgical
guides were developed, and the implant was manufactured using 3D printing in
titanium (Figures 5 and 6).
The
guides were designed with four points of support on
peripheral glenoid landmarks, considering retractor placement, to allow precise
positioning of the central guide pin or drill.
Prior
to final manufacturing, full scapular biomodels and
full-scale plastic prototypes of the baseplates were produced (Figure 7). This enabled the surgeon to become
familiar with the intraoperative scenario and, if satisfactory, approve the
final design.
The
definitive patient-specific baseplate was manufactured using 3D printing
technology in ELI grade 5 trabecular titanium, with a trabecular metal
interface of 0.4 mm thickness, 70% porosity, and an additional sandblasted
surface treatment to enhance osseointegration (Figures 8 and 9).
Surgical Technique
Patients
underwent surgery in the beach-chair position under sedation and regional
anesthesia of the affected limb. A deltopectoral
approach was used in all cases. After release of adhesions and resection of
fibrotic tissue, osteosynthesis hardware or prosthetic
components were removed as appropriate.
Following
circumferential release of the glenoid, bony landmarks were
identified for placement of the 3D-printed titanium guide, which was
used as a template to guide drilling for the central peg (Figure 10).
A
full-scale plastic prototype of the baseplate was used
to verify fit, followed by implantation of the definitive patient-specific
component (Figures 11 and 12).
The
humeral component was addressed as required (either as
a primary implantation, revision procedure, or retention of the existing
implant) depending on the case.
All
patients followed the same postoperative protocol: immobilization in a sling
for the first 6 weeks, combined with gentle passive range-of-motion exercises.
From week 6 onward, assisted active exercises were introduced to improve
mobility, followed by progressive strengthening beginning at week 12. Full
recovery was expected within 6 to 12 months.
RESULTS
The
series comprised eight consecutive patients with a minimum follow-up of 12
months. The mean age was 67.3 years (range 38-84). Seventy-five percent were
women and 25% were men. The affected shoulder was the right in six patients and
the left in the remaining two (Table 1).
The
mean operative time was 142.5 minutes (range 105-180) (Table
2).
The
distribution of bone defects according to the Gohlke
classification was as follows: type 3 (25%), type 4
(50%), and type 5 (25%).
In
this series, significant improvements were observed in all analyzed variables.
Forward elevation increased by a mean of 78.8° (95% confidence interval [CI]
65.0-92.5; p < 0.001), external rotation improved by 7.5° (95% CI 1.6-13.4;
p = 0.019), and internal rotation improved from the gluteal level (trochanter-L5)
to a mean of T12 (range T7-gluteal level). Pain, measured using the visual
analog scale, decreased by 6.4 points (95% CI-7.6 to -5.2; p < 0.001). The
Constant–Murley score increased by 48.9 points (95%
CI 42.6-55.1; p < 0.001) (Table 3).
Radiographic
analysis showed appropriate implant integration in seven of the eight cases (Figures 13 and 14).
In one
patient with sequelae of a previously treated proximal humeral fracture with a
PHILOS plate and severe glenoid erosion caused by the implant screws, loosening
of the glenoid baseplate with mild migration was observed.
This resulted in moderate functional limitation, although pain control remained
adequate. The patient was satisfied with pain relief and declined further
surgical intervention; therefore, conservative management was adopted (Figure 15).
DISCUSSION
Our
results demonstrate that the use of patient-specific glenoid implants is a
technically feasible option for the treatment of severe glenoid defects in
revision surgery. Despite the complexity of the included cases, only one
complication related to loosening of the glenoid component was observed,
highlighting the stability achieved in most patients over a minimum follow-up
of 12 months.
These
findings are consistent with previous studies, although differences exist in
patient populations and surgical indications. Chammaa
et al.,6
in a series of 37 patients, the largest reported to date, described
favorable outcomes at a mean follow-up of 60 ± 25 months, with significant
improvements in the Oxford Shoulder Score (from 11 to 27 points) and the
Subjective Shoulder Value (from 23% to 60%). Active elevation increased from
39° ± 23° to 64° ± 38°, and external rotation from 6° ± 16° to 15° ± 17°.
However,
their study focused on primary arthroplasties, whereas our series includes only
revision procedures, which reinforces the relevance of our findings by
demonstrating that patient-specific implants are also effective in more complex
scenarios. In addition, it is important to note that the implant used in the
study by Chammaa et al. was not patient-specific but
rather based on a design concept similar to that of hip arthroplasty, which may
limit its effectiveness compared with implants specifically designed for
shoulder anatomy.
Likewise,
Rangarajan et al. reported notable improvements in
the Constant-Murley score (from 24.6 to 60.4) and the
ASES score (from 32 to 79), as well as in range of motion, with increases in
forward elevation (from 53° to 124°), abduction (from 42° to 77°), and external
rotation (from 17° to 32°) in a series of 19 patients. However, that study
included both primary and revision arthroplasties, resulting in a more
heterogeneous cohort, which may limit direct comparability with our findings.
In the present series, patients had a mean of 1.3 prior surgeries (range 1-4),
reflecting a more complex clinical scenario; nevertheless, the functional
improvements observed were comparable.11
Bodendorfer
et al. and Ortmaier et al. reported similar outcomes
in terms of range of motion and function, with a minimum follow-up of 24
months, in series of 11 and 9 patients, respectively.2,4 In
the study by Bodendorfer et al., improvements were
observed in forward elevation (from 95° to 150°), external rotation (from 13°
to 40°), and internal rotation (from the sacrum to L3).4 Ortmaier et al. reported improvements in the Constant-Murley score (from 10.9 to 51.7), the Subjective Shoulder
Value (from 11% to 52%), and abduction (from 19° to 121°).2 It is
important to note that, in the series by Bodendorfer
et al., surgeries were performed by four surgeons across three different
institutions, which may have introduced variability in both technique and
surgical experience.4
In our
series, no intraoperative complications were observed. A single postoperative
complication related to the glenoid component was recorded: one patient with a
type 5 defect according to the Gohlke classification
developed signs of baseplate loosening during follow-up. Although
this rate (12.5%) is comparable to that reported in other series, studies such
as that by Chammaa et al. reported a complication
rate of 24% (9 of 37 patients), and Rangarajan et al.
reported a rate of 21% (4 of 19 patients), including infections, hematomas, and
intraoperative fractures.6,11 Other authors, such as Porcellini
et al., described minor radiographic findings and one dislocation in a series
of six patients,5 whereas Bodendorfer et al.,4 and Ortmaier et al.2 reported
no complications. This variability may be attributable to differences in
the defects treated, surgical experience, technique, and duration of follow-up.
Patient-specific
glenoid implants offer several important advantages. They allow precise
adaptation to complex bone defects, improving surgical accuracy and optimizing
implant fixation and primary stability. This is particularly relevant in
patients with severe defects in whom conventional implants may not provide an
adequate solution.2 Preoperative planning
using the described methodology enables accurate assessment of bone stock and
density, maximizing contact with the native glenoid and optimizing screw
positioning, trajectory, and length. This results in improved primary stability
and subsequent osseointegration, as observed in our
series. Furthermore, the ability to design patient-specific implants allows
effective management of anatomical variability and the specific characteristics
of each glenoid defect.
However,
this technique has certain drawbacks, such as a steep learning curve due to the
low incidence of cases, which may impact operative time. In our series,
variability in operative time was directly related to the type of procedure
performed; although all cases involved revision arthroplasty, in two patients
only the glenoid component was revised, whereas in the remaining cases both
components were addressed. In cases of failed osteosynthesis,
additional time required for hardware removal must be considered; similarly,
failed hemiarthroplasties required removal of the humeral component, generally
involving a humeral osteotomy. Another limitation is the delay between
acquisition of the 3D CT scan and surgery. In our experience, once preoperative
planning is approved, the implant can be manufactured and made available to the
surgeon within a minimum of 5 weeks. The cost of patient-specific implants is
higher than that of standard off-the-shelf implants, which may represent a limiting
factor in certain settings.
This
study has several limitations that should be considered. First, it is a
retrospective study without a control group, which may limit interpretation of
the results. Although the number of patients included is comparable to that of
published international studies, the small sample size (eight patients) limits
generalizability and may not capture the full variability of the population
with severe glenoid defects. This small sample also precludes statistical power
calculation, thereby limiting the validity of comparisons and introducing a
potential risk of Type II error. Therefore, the results should be interpreted
as preliminary and descriptive, and studies with larger sample sizes are
required to confirm the effectiveness of patient-specific implants. In
addition, the follow-up period was relatively short (12 months), precluding
adequate assessment of long-term implant survival.
Furthermore,
no independent evaluator was used to assess range of motion, as measurements were performed by the treating surgeon during
clinical follow-up visits. Another limitation is that imaging follow-up was
performed exclusively with plain radiographs in two projections (true
anteroposterior and scapular axial views). Given the complex three-dimensional
structure of these implants, CT evaluation could have provided more precise
information regarding osseointegration; however, it
was not performed due to additional costs and radiation exposure.
The
novelty of this technique and its limited indications also restrict the number
of patients treated with patient-specific glenoid implants to date, which
further limits generalizability. Longer follow-up and larger cohorts are
required to draw more robust conclusions regarding implant survival and complication
rates.
Despite
these limitations, this study has several notable strengths. To our knowledge,
it is the first study published in a national and Latin American setting
evaluating patient-specific glenoid implants with a minimum follow-up of one
year, thereby providing a valuable contribution. The homogeneity of the study
population is another important strength, as all patients underwent revision
arthroplasty, allowing comparison with similar series. In addition, all procedures were performed by a single surgeon at a single
institution, ensuring consistency in surgical technique and experience, and
minimizing variability in outcomes.
CONCLUSIONS
Our
study reports promising preliminary findings on the use of patient-specific
glenoid implants in revision reverse shoulder arthroplasty, demonstrating
positive outcomes in terms of range of motion, function, and pain reduction,
with a low complication rate. Although further studies with larger cohorts and
longer follow-up are required, these preliminary results support the use of
this technology in patients with severe glenoid defects.
REFERENCES
1. Burton R, Adam J, Holland
P, Rangan A. A review of custom implants for glenoid
bone deficiency in reverse shoulder arthroplasty. J Orthop 2023;36:65-71.
https://doi.org/10.1016/j.jor.2022.11.016
2. Ortmaier R, Wierer
G, Gruber MS. Functional and radiological outcomes after treatment with
custom-made glenoid components in revision reverse shoulder arthroplasty. J Clin Med 2022;11(3):551. https://doi.org/10.3390/jcm11030551
3. Thati B, Bodanki
C, Badam VK, Reddy MV, Reddy AVG. Custom 3D printed
jigs in salvage reverse shoulder arthroplasty for failed four-part proximal humerus fracture fixation: a case report. J Orthop Case Rep 2020;10(2):25-8. https://doi.org/10.13107/jocr.2020.v10.i02.1682
4. Bodendorfer BM, Loughran
GJ, Looney AM, Velott AT, Stein JA, Lutton DM, et al. Short-term outcomes of reverse shoulder
arthroplasty using a custom baseplate for severe glenoid deficiency. J Shoulder Elbow Surg
2021;30(5):1060-7. https://doi.org/10.1016/j.jse.2020.08.002
5. Porcellini G, Micheloni
GM, Tarallo L, Paladini P, Merolla G, Catani F. Custom-made reverse shoulder
arthroplasty for severe glenoid bone loss: review of the literature and our
preliminary results. J Orthop Traumatol 2021;22(1):2. https://doi.org/10.1186/s10195-020-00564-6
6. Chammaa R, Uri O, Lambert S.
Primary shoulder arthroplasty using a custom-made hip-inspired implant for the
treatment of advanced glenohumeral arthritis in the
presence of severe glenoid bone loss. J
Shoulder Elbow Surg 2017;26(1):101-7.
https://doi.org/10.1016/j.jse.2016.05.027
7. Murphy J, Todd E, Wright
MA, Murthi AM. Evaluation of clinical and radiographic
outcomes after total shoulder arthroplasty with inset Trabecular Metal-backed
glenoid. J Shoulder Elbow Surg 2022;31(5):971-7. https://doi.org/10.1016/j.jse.2021.10.014
8. Debeer P, Berghs
B, Pouliart N, Van den Bogaert
G, Verhaegen F, Nijs S. Treatment of severe glenoid
deficiencies in reverse shoulder arthroplasty: the Glenius
Glenoid Reconstruction System experience. J
Shoulder Elbow Surg 2019;28(8):1601-8.
https://doi.org/10.1016/j.jse.2018.11.061
9. Peri G, Troiano E, Colasanti GB, Mondanelli N,
Giannotti S. Custom-made glenoid baseplate and intra-operative navigation in
complex revision reverse shoulder arthroplasty: a case report. J Shoulder Elb Arthroplast 2024;8:1-6. https://doi.org/10.1177/17585732231225968
10. Rashid MS, Cunningham L,
Shields DW, Walton MJ, Monga P, Bale RS, et al.
Clinical and radiologic outcomes of Lima ProMade
custom 3D-printed glenoid components in primary and revision reverse total
shoulder arthroplasty with severe glenoid bone loss: a minimum 2-year
follow-up. J Shoulder Elbow Surg 2023;32(10):2017-26. https://doi.org/10.1016/j.jse.2023.04.020
11. Rangarajan R, Blout
CK, Patel VV, Bastian SA, Lee BK, Itamura JM. Early
results of reverse total shoulder arthroplasty using a patient-matched glenoid
implant for severe glenoid bone deficiency. J
Shoulder Elbow Surg 2020;29(7S):S139-48.
https://doi.org/10.1016/j.jse.2020.04.024
12. Gohlke F, Werner B. Humerale und glenoidale Knochendefekte in der Schulterendoprothetik:
Klassifikation und Behandlungsprinzipien.
Orthopade 2017;46(12):1008-14.
https://doi.org/10.1007/s00132-017-3484-5
D. J. Gómez ORCID ID: https://orcid.org/0000-0003-0258-6802
G.
M. Viollaz ORCID ID: https://orcid.org/0000-0002-4573-883X
G.
Teruya ORCID ID: https://orcid.org/0000-0001-7342-1859
Á.
Muratore ORCID ID: https://orcid.org/0000-0001-7540-7137
A.
Tedeschi ORCID ID: https://orcid.org/0000-0001-5704-3122
L. Gutiérrez ORCID ID: https://orcid.org/0009-0000-4603-313X
R. Durán ORCID ID: https://orcid.org/0000-0002-8789-3221
S. Ávila Posada ORCID ID: https://orcid.org/0009-0008-8035-3522
Daniel
Moya ORCID ID: https://orcid.org/0000-0003-1889-7699
Received on July 22nd,
2025. Accepted after evaluation on October 20th, 2025 • Dr. NADIA GABOTTO LOREDO • Loredogabottonadia@gmail.com
•
https://orcid.org/0009-0001-8122-0237
How to cite this article:
Gómez
DJ, Gabotto Loredo N,
Viollaz GM, Teruya G, Muratore Á, Tedeschi
A, et al. Custom Reverse Shoulder Arthroplasty for Severe Postoperative Glenoid
Bone Defects: A Retrospective Case Series. Rev
Asoc Argent Ortop Traumatol 2026;91(2):103-117. https://doi.org/10.15417/issn.1852-7434.2026.91.2.2199
Article Info
Identification: https://doi.org/10.15417/issn.1852-7434.2026.91.2.2199
Published: April, 2026
Conflict of interests:
The authors declare no conflicts of interest.
Copyright: © 2026, Revista de la Asociación Argentina de Ortopedia y Traumatología.
License: This
article is under Attribution-NonCommertial-ShareAlike 4.0 International Creative Commons License
(CC-BY-NC-SA 4.0).