Abstract
Purpose
Total hip arthroplasty (THA) in patients with acetabular protrusion can be challenging due to the lack of acetabular bone support. Some patients may require a major reconstructive procedure to aid in the fixation of the acetabular components. The purpose of the study was to characterize the type of acetabular defects in patients with protrusion and to assess the need for major reconstruction.
Methods
This was a retrospective study of 125 THAs performed at a single tertiary center from 2010 to 2020 by a single surgeon in patients with acetabular protrusion. Demographics, operative details and radiographic details were recorded. THAs in which a structural graft, cage or augment were used for acetabular reconstruction were considered to have undergone a major reconstructive procedure and a classification system was proposed to identify patients requiring major reconstruction.
Results
10 (8 %) THAs required a major reconstruction in our cohort (2- augment, 8- structural allograft. There was significant difference in acetabular defects based on the etiology. Major reconstruction was more common when the diagnosis was sequalae of tuberculosis or post-traumatic arthritis. Simple cavitatory defects seldom required a major reconstruction while combined defects and medial wall deficiency increased the risk of requiring a major reconstruction.
Conclusion
The severity of defects varied considerably with majority of patients having cavitatory type of defect which can be successfully managed without the need of any major reconstructive measures. A simple classification based on the type of defect and medial wall competency appeared to be useful in identifying patients who may need a major reconstructive procedure during THA.
Keywords: Total hip arthroplasty, Acetabular protrusion, Classification, Reconstruction
1. Background
Total hip arthroplasty (THA) is a highly successful procedure with excellent long term functional outcomes.1,2 With advancements in THA, the indications have expanded and many patients with severe forms of arthritis and bone defects are undergoing THA successfully. Acetabular protrusion or protrusio acetabuli refers to medial migration of the femoral head into the true pelvis.3 Acetabular protrusion can be idiopathic or secondary to inflammatory arthritis, infection, trauma, etc.4 THA in acetabular protrusion is a demanding procedure due to significant medial and proximal migration of the joint center, deficient medial bone and reduced peripheral bony support.5
Adequate acetabular bone support is usually required for a successful fixation of the acetabular component in THA. Many studies have described various techniques to perform THA in these settings such as use of bone grafts, large sized cups, augments, etc.6, 7, 8 Although, many centers have described their experience on treatment of THA in patients with acetabular protrusion, there is no general consensus regarding the use of adjuncts in THA for these patients.5,9,10 Moreover, currently there are no classifications that can help in the management of THA based on the acetabular defect seen in protrusion patients. While classifications like AAOS and Paprosky classifications are helpful in assessing the severity of the defect, they are not specific for protrusion patients and their utility in guiding treatment is unclear.11, 12, 13
Understanding the acetabular defects in patients with protrusion and classifying the defects may help in identifying the patients who require a major reconstructive measure during THA. Therefore, we conducted this study in an attempt to characterize and possibly classify the acetabular defects in protrusion, so that appropriate treatment decisions can be made. The objectives of the study were to 1) characterize the type of acetabular defects in patients undergoing primary THR with protrusion, and 2) to classify patients based on the need of major reconstruction.
2. Methods
This is a retrospective study of patients who underwent THA with acetabular protrusion from 2010 to 2020. Patients were identified from the database of THAs performed by single surgeon at a quaternary level hospital. Their medical records and imaging were reviewed by one or more of the authors to confirm the diagnosis. Clinical, and radiographic details were collected by manual review of the medical records.
All patients with protrusion undergoing primary THA were included. Acetabular protrusion was defined based on plain available radiographs when the femoral head migrated medial to Kohler's line (a line from the lateral border of the obturator foramen to the medial border of the sciatic notch).14 Both unilateral and bilateral cases were included. We did not include patients who were undergoing revision (N = 4), or those who did not have a follow-up of at-least 2 years. Demographic variables such as age, gender along with the indication of the procedure was recorded. We included 125 hips in 98 patients with an average follow-up of 5.5 ± 3.9 years (range, 2–12 years). The mean age was 45.9 ± 14.6 years (range, 13–85 years). There was 64 males (mean age: 42.1 ± 16.5, range: 13–85 years) and 61 females (mean age: 49.9 ± 11.0, range: 19–69 years). The choice of implant was based on the surgeon's choice. All patients received uncemented implants. Out of the 125 hips, in 23 (18 %) hips short stems were used. In our center, young patients (less than 40) are offered short stems as they can preserve bone and can be easily removed in case of a future revision.15
The images were reviewed independently by two authors and the following radiographic parameters were included: type of defect (cavitatory, segmental, combined), head migration, head destruction, and continuity of the medial wall. The type of defect was defined based on the study by Antonio et al.13 which is also used in the AAOS classification. A segmental deficiency is any complete loss of bone in the supporting architecture of the acetabulum. It can be peripheral or central (loss of medial wall). A cavitatory deficiency represents a volumetric loss of bone with an intact acetabular rim. Like segmental deficiencies, cavitatory deficiencies can be peripheral or central. Patients can have combined segmental and cavitatory deficiency as well. Head migration was classified into mild, moderate or severe based on the percentage of head crossing the Kohler's line (Mild: <25 %, Moderate: 25–50 %, Severe: >50 %). Head destruction was classified into: <50 %, 50–75 %, >75 %. This was assessed using an imaginary circle at the head on the AP films to estimate the amount of defect. Additionally, radiographs were also classified using both AAOS and Paprosky classification for acetabular defects.12,13
THAs in which a structural graft, cage or augment was used for acetabular reconstruction were considered to have undergone a major reconstructive procedure. This was because we believed that the use of any of these adjuncts deviates significantly from primary surgery. The decision for need of major reconstruction was anticipated based on the intra-operative characterization of the defects and the assessment of support of the acetabular implants by the senior surgeon. All patients were followed for a minimum of 2 years to look for clinical outcomes such as revision, infection, or dislocation, functional outcomes using Harris Hip Score and radiological outcomes. Categorical variables were compared using chi-squared test while continuous variables were compared using Wilcoxon rank sum test. Based on the radiographic parameters which correlated with major reconstruction, a simple graded classification system was devised. The frequency of major reconstructive procedures in each of the classification groups was assessed. Statistical analysis was performed using Stata statistical software, version 12 (StataCorp, College Station, TX).
3. Results
Ten (8 %) THAs required a major reconstruction in our cohort (2- augment, 8- structural allograft). The demographics of the cohort are detailed in Table 1. Patients requiring a major reconstruction were more likely to be males (p = 0.017). Major reconstruction was more common when the diagnosis was tubercular sequalae or post-traumatic arthritis (p < 0.001). Presence of combined cavitatory and segmental defect (p < 0.001), large head defect (p = 0.028), higher degree of head migration (p < 0.001), and incompetent medial wall (p = 0.044) were found to be associated with major reconstruction. Both the AAOS (p < 0.001) and Paprosky classification (p < 0.001) also were significantly different between the two groups.
Table 1.
Baseline characteristics of THAs included in the study.
| Variable | Major reconstruction (n = 10) | No major reconstruction (n = 115) | P-value |
|---|---|---|---|
| Age | 45.6 ± 16.2 | 45.9 ± 14.5 | 0.944 |
| Gender | 0.017 | ||
| Male | 9 (90 %) | 55 (47.8 %) | |
| Female | 1 (10 %) | 60 (52.2 %) | |
| Diagnosis | <0.001 | ||
| TB | 3 (30 %) | 22 (19.1 %) | |
| Inflammatory | 0 (0 %) | 71 (61.7 %) | |
| Post traumatic | 6 (60 %) | 5 (4.3 %) | |
| Others | 1 (10 %) | 17 (14.8 %) | |
| Side | 0.327 | ||
| Right | 7 (70 %) | 58 (50.4 %) | |
| Left | 3 (30 %) | 57 (49.6 %) | |
| Radiologic parameters | |||
| Defect type | <0.001 | ||
| avitatory | 0 (0 %) | 112 (95.7 %) | |
| Segmental + cavitatory | 10 (100 %) | 3 (2.6 %) | |
| Head defect | 0.028 | ||
| <50 % | 3 (30 %) | 79 (68.7 %) | |
| 50–75 % | 3 (30 %) | 21 (18.3 %) | |
| >75 % | 4 (40 %) | 15 (13 %) | |
| Medial Wall | 0.044 | ||
| Intact | 7 (70 %) | 107 (93 %) | |
| Non-continuous | 3 (30 %) | 8 (7 %) | |
| Head Migration | <0.001 | ||
| Mild | 1 (10 %) | 77 (67 %) | |
| Moderate | 3 (30 %) | 25 (21.7 %) | |
| Severe | 6 (60 %) | 13 (11.3 %) | |
| AAOS score | <0.001 | ||
| 2a | 1 (10 %) | 12 (10.4 %) | |
| 2 b | 0 (0 %) | 88 (76.5 %) | |
| 3 | 6 (60 %) | 3 (2.6 %) | |
| 4 | 2 (20 %) | 7 (6.1 %) | |
| 5 | 1 (10 %) | 5 (4.3 %) | |
| Paprosky | <0.001 | ||
| 2a | 0 (0 %) | 75 (65.2 %) | |
| 2 b | 3 (30 %) | 23 (20 %) | |
| 2c | 1 (10 %) | 4 (3.5 %) | |
| 3a | 4 (40 %) | 7 (6.1 %) | |
| 3 b | 2 (20 %) | 6 (5.2 %) | |
The most common etiology was inflammatory arthritis followed by tubercular sequalae. There was significant difference in acetabular defects based on the etiology (Table 2). Post-traumatic arthritis was more likely to result in an incompetent medial wall (p = 0.002). Both tubercular and post-traumatic arthritis patients were likely to have a combined acetabular defect (p < 0.001), and also had higher risk of major reconstruction. There was significant difference in AAOS classification based on etiology (p < 0.001), but not on Paprosky classification (p = 0.147).
Table 2.
Characteristics of acetabular defects based on etiology.
| Variable | TB (n = 25) | Inflammatory Disease (N = 71) | Others (N = 18) | Post traumatic (N = 11) | P-value |
|---|---|---|---|---|---|
| AAOS score | <0.001 | ||||
| 2a | 3 (12 %) | 7 (9.9 %) | 3 (16.7 %) | 0 (0 %) | |
| 2 b | 17 (68 %) | 55 (77.5 %) | 14 (77.8 %) | 2 (18.2 %) | |
| 3 | 4 (16 %) | 2 (2.8 %) | 0 (0 %) | 3 (27.3 %) | |
| 4 | 1 (4 %) | 2 (2.8 %) | 1 (5.6 %) | 5 (45.5 %) | |
| 5 | 0 (0 %) | 5 (7 %) | 0 (0 %) | 1 (9.1 %) | |
| Paprosky | 0.147 | ||||
| 2a | 15 (60 %) | 45 (63.4 %) | 12 (66.7 %) | 3 (27.3 %) | |
| 2 b | 6 (24 %) | 15 (21.1 %) | 4 (22.2 %) | 1 (9.1 %) | |
| 2c | 0 (0 %) | 3 (4.2 %) | 0 (0 %) | 2 (18.2 %) | |
| 3a | 2 (8 %) | 5 (7 %) | 1 (5.6 %) | 3 (27.3 %) | |
| 3 b | 2 (8 %) | 3 (4.2 %) | 1 (5.6 %) | 2 (18.2 %) | |
| Defect type | <0.001 | ||||
| Cavitatory | 21 (84 %) | 69 (97.2 %) | 17 (94.4 %) | 5 (45.5 %) | |
| Segmental + cavitatory | 4 (16 %) | 2 (2.8 %) | 1 (5.6 %) | 6 (54.5 %) | |
| Head defect | 0.743 | ||||
| <50 % | 14 (56 %) | 49 (69 %) | 12 (66.7 %) | 7 (63.6 %) | |
| 50–75 % | 6 (24 %) | 13 (18.3 %) | 4 (22.2 %) | 1 (9.1 %) | |
| >75 % | 5 (20 %) | 9 (12.7 %) | 2 (11.1 %) | 3 (27.3 %) | |
| Medial Wall | 0.002 | ||||
| Intact | 23 (92 %) | 68 (95.8 %) | 17 (94.4 %) | 6 (54.5 %) | |
| Non-continuous | 2 (8 %) | 3 (4.2 %) | 1 (5.6 %) | 5 (45.5 %) | |
| Head Migration | 0.284 | ||||
| Mild | 15 (60 %) | 46 (64.8 %) | 12 (66.7 %) | 5 (45.5 %) | |
| Moderate | 6 (24 %) | 17 (23.9 %) | 4 (22.2 %) | 1 (9.1 %) | |
| Severe | 4 (16 %) | 8 (11.3 %) | 2 (11.1 %) | 5 (45.5 %) |
While multiple radiographic parameters were found to be associated with need of major reconstruction, we developed a simple classification using just two parameters: the type of defect and medial wall competency (Table 3). Simple cavitatory defects seldom required a major reconstruction (Fig. 1). But, combined defects and the presence of a medial wall defect increased the risk of using a major reconstruction. The various treatment options that can be used based on the classification stage are also provided in Table 3.
Table 3.
Proposed classification of acetabular defects in protrusion patients in guiding treatment.
| Type | Description | Frequency of Major Reconstruction | Treatment Options |
|---|---|---|---|
| Type 1 | Isolated cavitatory defect | 0 % (0/112) |
|
| Type 2a | Cavitatory with segmental defect | 70 % (7/10) |
|
| Type 2 b | Cavitatory with segmental defect with incompetent medial wall | 100 % (3/3) |
|
Fig. 1.
Preoperative radiographs (left panel) of a patient with bilateral arthritis secondary to inflammatory disease with protrusion in the left side (Type 1). Post-operative radiographs (right panel) showing bilateral THA without the need of any major reconstruction.
In our study cohort, only two patients had a revision (one for infection, one for dislocation) during minimum of two-year follow-up. The mean HHS recorded at the final follow-up was 87 (range 77–97). 102 (82 %) patients were walking independently while the remaining 33 (18 %) patients were walking with a support. At the latest follow-up, all the cups were well integrated. There was no radiolucency between the cup and socket in any patient indicating incorporation of the morselized bone graft. None of the patients had any osteolysis or evidence of component migration in the follow-up x-rays. There was no evidence of femoral stem subsidence or loosening. Heterotopic ossification was observed in 3 patients.
4. Discussion
Understanding the acetabular defects is important to anticipate challenges during total hip replacement in patients with protrusion. The currently available classification systems are either not specific for patients with protrusion or are of limited utility in guiding management. We conducted a retrospective study of 125 primary THAs done in patients with protrusion at a single center and found that majority of the patients had minimal defects not requiring any major reconstruction. A simple classification based on the type of defect and medial wall competency appeared to be useful in identifying patients who may need a major reconstructive procedure during total hip replacement.
There are a number of limitations for this study. Firstly, this was retrospective study and is subject to the limitations of the same. Acetabular defects were assessed using simple radiographs which were digitally stored by the surgeon in his database. Some of these x-rays were not digital and did not have scales which prevented quantification of the protrusion and grading using Sotelo-Garza classification.16 We also did not use pre-operative CT scans for quantification of acetabular defects which may have improved the accuracy of characterization of the defects. The proposed classification was based on data of the study and has not been externally validated. Only clinical outcomes were assessed in this study and functional outcomes which are important to understand the quality of life of patients after THA were not assessed in this study. Finally, this was a single center single surgeon study with limited size, and the results may not be generalizable to other centers or patient cohorts. For example, pathological conditions like tubercular arthritis and end stage inflammatory arthritis are less frequently encountered in western countries.
Isolated cavitatory defect was the most commonly encountered type of defect. In our cohort, we did not find any patients with isolated segmental defects. This might be because isolated loss of acetabular rim is unlikely to cause protrusion of the femoral head. Our study also found that the type of defect was largely determined by the etiology. In the present study, majority of the patients suffered from inflammatory arthritis followed by infection. In a study by Lee et al.17 on 26 hips, inflammatory arthritis (54 %) was the most common cause followed by post traumatic arthritis (27 %). Other studies has also reported similar results.18,19 However, very few studies have reported on protrusion in patients with tubercular sequalae. The large number of patients with post tubercular arthritis may be unique to the Indian population where tuberculosis is endemic. Compared to inflammatory arthritis, the defect type was more severe in tubercular patients. Also, these patients had a more severe femoral head loss which may prevent use of femoral head as structural autografts in patients requiring additional bone support. Our study also found that post-traumatic patients had a higher incidence of medial wall defects. Prior trauma and associated fracture of the acetabulum can result in a weak quadrilateral plate which can result in protrusion.20
Our study found that majority of the patients could be managed without a need of major reconstruction. Impaction grafting was used in all patients in this study using graft obtained from the reamed acetabular socket and remaining femoral head. This suggests that THR can be performed without additional technical challenges in most cases of protrusion. Use of cementless cups have become a standard these days, and often screws are inserted for additional stability. Most of the protrusions have a deep socket with an intact rim. In these patients conventional THA techniques can provide excellent fixation of the cup as there is sufficient bone stock. Even in patients with severe protrusion, the acetabular rim is often intact as a result of which, they can be managed successfully with conventional techniques supplemented with impaction bone grafts. Rosenber et al.9 reported a 90 % survival rate using morselized bone grafts in a study of 31 THAs with protrusion.
We also proposed a simple classification which can be easily applied with minimal expertise and time. Based on the findings of our study, we believe that the type of defect is more important than the severity of the protrusion. Even in patients with severe medial migration of the cup, an additional reconstructive option may not be required. However, it is important to note that maintaining the centre of rotation of the hip might be challenging and pre-operative assessment of offsets and use of templates can aid surgeon in maintaining a biomechanically suitable position. Our results also suggest that patients with combined defects, especially when associated with medial wall discontinuity (often seen in post-traumatic cases), have a very high risk of major reconstruction. In such challenging cases, surgeons should be prepared with reconstructive options to establish adequate fixation of the cup (Table 3).
In summary, our study found that inflammatory arthritis and post-infective sequalae are the major causes for protrusion in patients in the cohort. The severity of defects varied considerably with majority of patients having cavitatory type of defect with minimal damage to the acetabular rim enabling THA using conventional techniques. Using a simple classification tool can aid surgeons in planning THA in patients with acetabular protrusion. Further studies are required to assess the long term clinical and functional outcomes in these patients.
CRediT authorship contribution statement
Rajesh Malhotra: study design, study oversight, manuscript editing. Deepak Gautam: study design, data collection, analysis, manuscript writing. Jaiben George: study design, data collection, analysis, manuscript writing.
Informed consent
Informed consent was obtained from all individual participants included in the study.
Consent for publication
The authors provide consent for publication of the study in International Orthopedics.
Funding
There is no funding source.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
Nil.
Contributor Information
Rajesh Malhotra, Email: rmalhotra62@gmail.com.
Deepak Gautam, Email: cmcdeepak@yahoo.com.
Jaiben George, Email: jaibengeorge@gmail.com.
References
- 1.Jones C.A., Pohar S. Health-related quality of life after total joint arthroplasty. Clin Geriatr Med. 2012;28(3):395–429. doi: 10.1016/j.cger.2012.06.001. [DOI] [PubMed] [Google Scholar]
- 2.Hooper G.J., Rothwell A.G., Stringer M., Frampton C. Revision following cemented and uncemented primary total hip replacement: a seven-year analysis from the New Zealand Joint Registry. J Bone Joint Surg Br. 2009;91(4):451–458. doi: 10.1302/0301-620X.91B4.21363. [DOI] [PubMed] [Google Scholar]
- 3.McBride M.T., Muldoon M.P., Santore R.F., Trousdale R.T., Wenger D.R. Protrusio acetabuli: diagnosis and treatment. J Am Acad Orthop Surg. 2001;9(2):79–88. doi: 10.5435/00124635-200103000-00002. [DOI] [PubMed] [Google Scholar]
- 4.Leunig M., Nho S.J., Turchetto L., Ganz R. Protrusio acetabuli: new insights and experience with joint preservation. Clin Orthop Relat Res. 2009;467(9):2241–2250. doi: 10.1007/S11999-009-0853-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Garcia-Cimbrelo E., Diaz-Martin A., Madero R., Munera L. Loosening of the cup after low-friction arthroplasty in patients with acetabular protrusion. The importance of the position of the cup. J Bone Joint Surg Br. 2000;82(1):108–115. doi: 10.1302/0301-620x.82b1.9796. [DOI] [PubMed] [Google Scholar]
- 6.Gautam D, Gupta S, Malhotra R. Total hip arthroplasty in acetabular fractures. J Clin Orthop trauma. 11(6):1090-1098. doi:10.1016/j.jcot.2020.10.037. [DOI] [PMC free article] [PubMed]
- 7.Dua A, Kiran K, Malhotra R, Bhan S. Acetabular reconstruction using fresh frozen bone allograft. Hip Int. 20(2):143-149. doi:10.1177/112070001002000203. [DOI] [PubMed]
- 8.Mullaji A.B., Shetty G.M. Acetabular protrusio: surgical technique of dealing with a problem in depth. Bone Joint Lett J. 2013;95-B(11 Suppl A):37–40. doi: 10.1302/0301-620X.95B11.32900. [DOI] [PubMed] [Google Scholar]
- 9.Rosenberg W.W., Schreurs B.W., de Waal Malefijt M.C., Veth R.P., Slooff T.J. Impacted morsellized bone grafting and cemented primary total hip arthroplasty for acetabular protrusion in patients with rheumatoid arthritis: an 8- to 18-year follow-up study of 36 hips. Acta Orthop Scand. 2000;71(2):143–146. doi: 10.1080/000164700317413102. [DOI] [PubMed] [Google Scholar]
- 10.Tauber C., Itzhak I.B., Malkin C. Bone grafting in total hip replacement for acetabular protrusion. A review of 11 operations. Acta Orthop Scand. 1984;55(5):507–509. doi: 10.3109/17453678408992947. [DOI] [PubMed] [Google Scholar]
- 11.Gozzard C., Blom A., Taylor A., Smith E., Learmonth I. A comparison of the reliability and validity of bone stock loss classification systems used for revision hip surgery. J Arthroplasty. 2003;18(5):638–642. doi: 10.1016/s0883-5403(03)00107-4. [DOI] [PubMed] [Google Scholar]
- 12.Telleria J.J.M., Gee A.O. Classifications in brief: Paprosky classification of acetabular bone loss. Clin Orthop Relat Res. 2013;471(11):3725–3730. doi: 10.1007/s11999-013-3264-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.D'Antonio J.A., Capello W.N., Borden L.S., et al. Classification and management of acetabular abnormalities in total hip arthroplasty. Clin Orthop Relat Res. 1989;243:126–137. [PubMed] [Google Scholar]
- 14.Berend M.E. Acetabular protrusio: a problem in depth. Orthopedics. 2008;31(9):895–896. doi: 10.3928/01477447-20080901-03. [DOI] [PubMed] [Google Scholar]
- 15.Malhotra R., Kumar V. Mid-term outcome of total hip arthroplasty using a short stem. J Orthop Surg. 2016;24(3):323–327. doi: 10.1177/1602400310. [DOI] [PubMed] [Google Scholar]
- 16.Sotelo-Garza A., Charnley J. The results of Charnley arthroplasty of hip performed for protrusio acetabuli. Clin Orthop Relat Res. 1978;132:12–18. [PubMed] [Google Scholar]
- 17.Lee B.S., Kim H.S., Kwon O.S., Lee Y.-K., Ha Y.-C., Koo K.-H. Is restoration of hip center mandatory for total hip arthroplasty of protrusio acetabuli? Hip pelvis. 2022;34(2):106–114. doi: 10.5371/hp.2022.34.2.106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Zhen P., Li X., Zhou S., Lu H., Chen H., Liu J. Total hip arthroplasty to treat acetabular protrusions secondary to rheumatoid arthritis. J Orthop Surg Res. 2018;13(1):92. doi: 10.1186/s13018-018-0809-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kinzinger P.J., Karthaus R.P., Slooff T.J. Bone grafting for acetabular protrusion in hip arthroplasty. 27 cases of rheumatoid arthritis followed for 2-8 years. Acta Orthop Scand. 1991;62(2):110–112. doi: 10.3109/17453679108999234. [DOI] [PubMed] [Google Scholar]
- 20.Qureshi A.A., Archdeacon M.T., Jenkins M.A., Infante A., DiPasquale T., Bolhofner B.R. Infrapectineal plating for acetabular fractures: a technical adjunct to internal fixation. J Orthop Trauma. 2004;18(3):175–178. doi: 10.1097/00005131-200403000-00009. [DOI] [PubMed] [Google Scholar]