Computed tomography osteoabsorptiometry is reliable for the determination of the subchondral bone mineralization distribution in the rabbit knee.

Rabbits are among the most frequently used animals in osteoarthritis research. It is meanwhile accepted that the subchondral bone plate (SBP) plays a key role in the development of osteoarthritis. The most suitable technique for analyzing subchondral bone mineralization is computed tomography osteoabsorptiometry (CT-OAM). Because CT-OAM has not yet been applied to smaller animals, the purpose of this study is to test the reliability of CT-OAM in the rabbit knee. Another important task in animal experiments is the intra- and interindividual difference of the measurement parameters. Our hypothesis is that there is no difference regarding both the position of the density maxima and the bone mineral density (BMD) of the SBP comparing right and left tibial plateaus of rabbits. For evaluating the reliability, a rabbit knee was examined by computed tomography 6 times at weekly intervals. The subchondral mineralization distribution was measured by means of CT-OAM. Positions of the density maxima and BMD of the SBP were determined in a standardized procedure. Furthermore, both parameters were evaluated in 6 female White New Zealand rabbits. Positions of density maxima and BMD in the SBP in left tibial plateaus were compared with right tibial plateaus. The relative coefficient of variation as a parameter for reproducibility was 1.6% for determining the position of the density maxima and 1.2% for measuring the BMD. The positions of density maxima and relative BMD between right and left tibial plateaus varied only about 2% intraindividually, whereas interindividual variance was about 10%. In conclusion, determination of the position of density maxima as well as BMD of the SBP by means of CT-OAM is reliable and reproducible in the rabbit knee. We recommend using the contralateral limb as control, because intraindividual accordance of the mineralization patterns and of the BMD of the SBP was higher than interindividual accordance.

An evaluation of the efficiency of the use of an anatomical third generation shoulder prosthesis in 102 patients.

PURPOSE OF THE STUDY: In this study a series of 102 cases was reviewed in which a so called third generation shoulder prosthesis had been used. There was an interest in evaluating the quality and efficiency of the outcome. MATERIAL AND METHODS: The patient population consisted of 32 men and 70 women. The mean age was 65.8 years. The mean postoperative follow-up time was 44.5 months. Indication for arthroplasty was all common degenerative and traumatic conditions of the shoulder. All patients were evaluated and analysed prospectively by a standardised protocol. The postoperative investigation consisted of a clinical examination and a radiographic analysis. We used the Constant and the Wülker Score. RESULTS: All patients had a significant improvement in shoulder function independent of the specific indication. The mean postoperative shoulder function attained 88% as measured by the Constant score. With regard to the different indications, the significant improvement was confirmed in each of the groups. Osteoarthritis reached the highest (91%) and rheumatoid arthritis the lowest value (71%). The results of total shoulder arthroplasties was better than in hemi-shoulder arthroplasties (93% versus 86%). CONCLUSIONS: In summary, it could be demonstrated that most indications for shoulder arthroplasty can be successfully treated by using a third generation implant. Total shoulders reach better results in a mid-term follow-up.

The effect of closed- and open-wedge high tibial osteotomy on tibial slope: a retrospective radiological review of 120 cases.

Radiographs of 110 patients who had undergone 120 high tibial osteotomies (60 closed-wedge, 60 open-wedge) were assessed for posterior tibial slope before and after operation, and before removal of the hardware. In the closed-wedge group the mean slope was 5.7 degrees (SD 3.8) before and 2.4 degrees (SD 3.9) immediately after operation, and 2.4 degrees (SD 3.4) before removal of the hardware. In the open-wedge group, these values were 5.0 degrees (SD 3.7), 7.7 degrees (SD 4.3) and 8.1 degrees (SD 3.9) respectively, when stabilised with a non-locking plate, and 7.7 degrees (SD 3.5), 9.4 degrees (SD 4.1) and 9.1 degrees (SD 3.8), when stabilised with a locking plate. The reduction in slope (-2.7 degrees (SD 4.1)) in the closed-wedge group and the increase (+2.5 degrees (SD 3.4), in the open-wedge group was significantly different before and after operation (p = 0.002, p = 0.003). In no group were the changes in slope directly after operation and before removal of the hardware significant (p > 0.05). There was no correlation between the amount of correction in the frontal plane and the post-operative change in slope. Posterior tibial slope decreases after closed-wedge high tibial osteotomy and increases after an open-wedge procedure because of the geometry of the proximal tibia. The changes in the slope are stable over time, emphasising the influence of the operative procedure rather than of the implant.

[Pathologic glenohumeral stress distribution in calcific tendinosis of the supraspinatus]. / Glenohumerale Fehlbeanspruchung bei Tendinosis calcarea der Supraspinatussehne.

AIM: Based on hypothesis that calcific tendonitis of the supraspinatus tendon (CTSSP) could be associated with glenohumeral imbalance, glenohumeral stress distribution was analyzed. METHODS: 26 patient shoulders with CTSSP, unsuccessfully treated by non-operative measures, were examined. A group of 26 macroscopically normal shoulder specimens served as controls. Analysis of glenohumeral stress distribution was indirect evaluating glenoid subchondral bone mineralization by computed tomography osteoabsorptiometry. Density distribution of glenoid subchondral bone mineralization and the position of the two most frequent density maxima were analyzed. RESULTS: Patterns of subchondral mineralization and position of the anterior density maximum were significantly different between both groups. CTSSP mostly presented with a monocentric, anteriorly increased mineralization indicative for a regional increase of stress. The inferior shift of the anterior density maximum demonstrates a parallel shift of glenohumeral stress distribution. CONCLUSION: Mineralization patterns indicate that glenohumeral stress distribution is not physiologic in CTSSP. Moreover, it is comparable with glenohumeral stress distribution as observed in atraumatic antero-inferior glenohumeral instability. Relevance of this observation should be proven for etiology of CTSSP.

Coracoid tip position on frontal radiographs of the shoulder: a predictor of common shoulder pathologies?

The coracoacromial arch is a static anterior-superior stabilizer of the humeral head. Thus coracoacromial arch inclination, which varies depending on coracoid tip position, may be related to shoulder pathologies. Therefore, we retrospectively analysed coracoid tip positions in the true anterior posterior view of different shoulder pathologies: reference shoulders (n=27), shoulders with rotator cuff tear (supraspinatus tear n=29; subscapularis tear n=21) and shoulders with anterior glenohumeral instability (traumatic n=17; atraumatic n=14). In supraspinatus tear shoulders, the coracoid tip projected onto inferior glenoid half in 86% of cases (type I coracoid), extending more inferiorly compared with reference group (p=0.0002) or subscapularis tear shoulders (p<0.0001). In contrast, 78% of cases with subscapularis tear show the coracoid tip projected onto the superior glenoid half (type II coracoid). Atraumatic glenohumeral instabilities had a more superior coracoid tip position than traumatic instabilities (p=0.04), but no differences were observed on basis of coracoid type or in comparison with normal controls. We conclude that coracoid tip position is highly variable. Since type I coracoids are predominant in shoulders with supraspinatus tears and type II coracoids in shoulders with subscapularis tears, coracoid tip position may thus provide a simple diagnostic complement for a probable site of rotator cuff tears.

Fluorescent microspheres are reliable for serial bone blood flow measurements.

The fluorescent microsphere method is one of the current techniques to determine regional blood flow in various organs. The purpose of this study was to examine the suitability of fluorescent microspheres for serial measurement of regional bone blood flow. Six anesthetized female New Zealand rabbits received five left ventricular injections of fluorescent microspheres in 20-minute intervals. To test the precision of the measurement two types of fluorescent microspheres were injected simultaneously at the first and last injections. Blood flow was calculated in the kidneys, lungs, brain, femurs, and tibias after measuring the fluorescence intensity in each reference blood and tissue sample. Comparison of blood-flow values obtained by simultaneously injected microspheres showed an excellent correlation and a minimal percentage difference at the first and last injections, indicating valid measurements of regional bone blood flow. No significant differences were observed when comparing blood flow in the corresponding regions of bones on the right side and left side. Mean blood flow in the femur and tibia significantly increased at the fourth injection whereas flow distribution within the femur and tibia essentially remained unchanged throughout the experiment. Comparison of blood flow values obtained by simultaneously injected microspheres showed moderate agreement for the kidneys and lungs at the last injections. Because this finding might be attributable to disturbances of microcirculation caused by accumulation of spheres in high-flow organs, the increase in regional bone blood flow observed in our experiments has to be interpreted carefully. This study showed that bone blood flow can be determined reliably in anesthetized rabbits by as many as three serial injections of fluorescent microspheres.

Medium to long-term results with the cementless threaded Acetabular Munich II type cup.

Compared to pressfit cups, little information exists about the results of screwed cups in hip arthroplasty. 51 cementless or hybrid (cemented stem) primary total hip replacements with a cementless corundium blasted titaniumn alloy threaded Aesculap Munich II type cup were examined with a mean follow up of 7.9 years. 23 of the patients were male and 28 patients were female. From these cups 22 were implanted on the right side and 29 on the left. The results were compared to 53 patients (28 male, 25 female, 29 right side, 24 left side) with the threaded Aesculap Munich I type cup, that has a smooth surface and a direct contact of bone with the polyethylene inlay. The mean follow up ws 10.2 years. The early and medium to long-term clinical and radiographic results show an encouraging improvement of the Merle d'Aubigné Score of the type II cup compared to the type I cup. Two of the Aesculap type Munich II cups had to be revised; four showed radiologic signs of loosening. In conclusion, the threaded Aesculap cup type Munich II seems to be a decisive advance in the development of threaded acetabular hip cups. The intermediate results exceed those from smooth-surface screwed rings and compare favourably with those from cemented cups and with those from cementless press-fit metal-backed cups.

Microspheres accurately predict regional bone blood flow.

Even though the microsphere method frequently is used to determinate bone blood flow, validation of this technique for bone blood flow measurement is incomplete. The method is based on the principle that injected microspheres are distributed with the arterial blood and trapped in the capillaries because of their diameter (15 microm). The number of spheres lodged in an organ is proportional to its blood flow. The number of radioactive or fluorescent microspheres in a specific organ is determined indirectly by measuring radioactivity or fluorescence intensity in the organ. In this study the reliability and precision of the microsphere method for determining bone blood flow was established using radioactive and fluorescent microspheres. Six female, anesthetized New Zealand rabbits received left ventricular injections of pairs of fluorescent and/or radioactive microspheres. The humerus, femur, and tibia were dissected in a standardized manner and blood flow was determined in each sample. Comparison of relative blood flow values showed an excellent correlation between radioactive and fluorescent microspheres. The percentage difference and variation between two simultaneously injected sets of microspheres was minimal for radioactive microspheres (0.8% +/- 9.6%) and for fluorescent microspheres (0.2% +/- 11.4%). Regional bone blood flow in different regions of the femur, tibia, or humerus ranged from 2.2-28.1 mL/minute/100 g, but there was no significant difference between right and left bone samples of the same region after repeated measurement. Radioactive and fluorescent microspheres allow precise determination of regional bone blood flow.

[The architecture of the subacromial space after full thickness supraspinatus tears]. / Die Architektur des subakromialen Raumes bei Supraspinatussehnenruptur.

AIM: The etiology of rotator cuff tears is multifactorial. An important factor is the damage of the rotator cuff by narrowing of the subacromial space. The purpose of this investigation was to estimate the influence of various metrical parameters on the size of the subacromial space. METHOD: We investigated 161 human macerated scapulae, 36 of them had a known tear of the supraspinatus tendon. All scapulae were photographed in two standard positions from the front and the lateral side. Defined distances and angles were measured using an image analyzing system followed by statistical analysis. RESULTS: Shoulders with a tear of the supraspinatus tendon showed a trend towards higher incidences of hooked acromions. Furthermore we found a significant higher incidence of elongated oval shaped glenoids in the group with supraspinatus tendon tear. In comparison to normal shoulders a significantly smaller distance from the top of the glenoid to the tip of the acromion and a greater distance from the top of the glenoid to the tip of the coracoid process was measured. In addition there was a significantly smaller coracoid angle and a smaller glenoid-spinal angle in this group. CONCLUSION: The width of the subacromial space depends on various parameters. Our data suggest that besides the acromion type the shape of the coracoid, the acromial angle, the spine-scapula angle and the cavitas-spine angle should be taken into account for diagnostic and therapeutic decisions.

Fluorescent microsphere method is suitable for chronic bone blood flow measurement: a long-term study after meniscectomy in rabbits.

The fluorescent microsphere (FM) method is considered a reliable technique to determine regional bone blood flow (RBBF) in acute experiments. In this study, we verified the accuracy and validity of this technique for measurement of RBBF in a long-term experiment and examined RBBF after meniscectomy. Twenty-four anesthetized female New Zealand white rabbits (3 groups, each n = 8) received consecutive left ventricular injections of FM in defined time intervals after meniscectomy: group 1 from preoperation to 3 wk postoperation; group 2 from 3 to 7 wk postoperation; and group 3 from 7 to 11 wk postoperation. To test the precision of the FM method, two FM species were injected simultaneously at the first and last measurement. After the experiment, humeri, femora, tibiae, and reference organs (kidney, lung, brain) were removed and dissected according to standardized protocols. Fluorescence was determined in each reference blood and tissue sample, and blood flow values were calculated. Blood flow in kidney, lung, and brain revealed no significant difference between right and left side and remained unchanged during the observation period, thus excluding errors due to shunting and dislodging of spheres in our experiments. Comparison of relative bone blood flow values obtained by simultaneously injected FM showed an excellent correlation at the first and last injection, indicating valid RBBF measurements in long-term experiments. We found a significant increase in RBBF 3 wk after meniscectomy in the right tibial condyles compared with the nonoperated left side. Similar changes were found in the femoral condyles. RBBF in other regions of tibia, femur, and humerus revealed no significant differences between right- and left-sided bone samples of the same region. Our results demonstrate that the FM method is valid for measuring RBBF in long-term experiments. In addition, we were able to demonstrate that meniscectomy leads to an increase in RBBF in the tibial condyles at a very early stage. This increase might be caused by stress-induced alterations of the subchondral bone.

[Dynamic palmar instability of the ulnar head]. / Die dynamische, palmare Ellenkopfinstabilität - Ein Fallbericht -

Despite obvious clinical symptoms, dynamic, palmar instability of the ulnar head remains a diagnostic and therapeutic problem. This is demonstrated by a case of chronic, palmar instability of the ulnar head after a tear of the triangular fibrocartilage complex (TFCC). A concomitant, minimal dorsal angulation of the radius after a fracture in childhood and a general capsulo-ligamentous laxity possibly were predisposing factors. Aspects of diagnostic and therapeutic options are discussed. In our case, due to secondary arthrosis, an arthrodesis of the distal radioulnar joint combined with a stabilization of the TFCC was opted for. Although the patient is painfree and mobile, she complains of a loss of power in supination. This salvage procedure probably could have been avoided by an earlier diagnosis.

Direct assessment and distribution of regional portal blood flow in the pig by means of fluorescent microspheres.

Measurement of regional organ blood flow by means of fluorescent microspheres (FM) is an accepted method. However, determination of regional portal blood flow (RPBF) cannot be performed by microspheres owing to the entrapment of the spheres in the upstream capillary bed of the splanchnic organs. We hypothesized that an adequate experimental setting would enable us to measure RPBF by means of FM and to analyze its distribution within the pig liver. A mixing chamber for the injection of FM was developed, and its capability to distribute FM homogeneously in the blood was evaluated in vitro. The chamber was implanted into the portal vein of six anesthetized pigs (23.5 +/- 2.9 kg body wt). Three consecutive, simultaneous injections of FM of two different colors into the chamber were performed. Reference portal blood samples were collected by means of a Harvard pump. At the end of the experiment, the liver was explanted and fixed in formalin before dissection. FM were isolated from the tissue samples by an automated process, and fluorescence intensity was determined. Comparison of 5,458 single RPBF values, determined by simultaneously injected FM, revealed good agreement (bias 2.5%, precision 12.7%) and high correlation (r = 0.97, r2 = 0,95, slope = 1.04, intercept = 0.05). Median RPBF was 1.07 +/- 0.78 ml x min(-1) x g(-1). Allocation of the blood flow values to the anatomic regions of the liver revealed a significantly higher RPBF (P = 0.01) in the liver tissue located close to the diaphragm compared with the rest of the organ and a significantly lower RPBF (P = 0.01) in the left liver lobe compared with the median and right lobes. The results show that the model presented makes it possible to measure RPBF by means of FM reliably and that RPBF is distributed heterogeneously in the porcine liver.

Determination of regional bone blood flow by means of fluorescent microspheres using an automated sample-processing procedure.

The determination of regional blood flow utilizing fluorescent microspheres (FMs) is an established method for numerous organs. Recent progress, in particular the automation of sample processing, has further improved this method. However, the FM method (reference sample technique), which allows repetitive measurement of regional organ blood flow, has so far not been used for the determination of blood flow in bone. The aim of the present study was to establish FM for the quantification of regional bone blood flow (RBBF). Female, anesthetized New Zealand rabbits (n = 6) received left ventricular injections of different amounts of FM at six subsequent time points. In order to examine the precision of RBBF determination, two different FM species were injected simultaneously at the sixth injection. At the end of the experiments the femoral and tibial condyles of each hind limb were removed and the fluorescence intensity in the tissue samples was measured by an automated procedure. In an in vitro study we have shown that acid digestion of the crystalline matrix has no effect on the fluorescence characteristics of FM. The determination of the number of spheres per tissue sample revealed that depending on the tissue sample size up to 3 x 10(6) spheres/injection were necessary to obtain about 400 microspheres in the individual bone samples. RBBF values of the tibial and femoral condyles did not differ at various injection intervals. The tibial blood flow values varied between 6.6 +/- 1.1 and 8.5 +/- 1.4 ml/min/100 g and were significantly higher than those of the femur (4.3 +/- 1.1 to 6.0 +/- 1.8 ml/min/100 g). The bone blood flow values obtained by simultaneous injection of two FM species correlated significantly (r = 0.96, slope = 1.06, intercept = 0.05), the mean difference was 0.39 +/- 1.11 ml/min/100 g. Our data demonstrate that the measurement of RBBF by means of FM allows a valid determination of RBBF.

Validity of fluorescent microspheres method for bone blood flow measurement during intentional arterial hypotension.

In this study, we compared bone blood flow values obtained by simultaneously injected fluorescent (FM) and radiolabeled microspheres (RM) at stepwise reduced arterial blood pressure. Ten anesthetized female New Zealand White rabbits received simultaneous left ventricular injections of FM and RM at 90, 70, and 50 mmHg mean arterial blood pressure (MAP). After the experiments, both kidneys and long bones of all four limbs were removed and dissected in a standardized manner. Radioactivity (corrected for decay, background, and spillover) and fluorescence were determined, and blood flow values were calculated. Relative blood flow values estimated for each bone sample by RM and FM were significantly correlated (r = 0.98, slope = 0.99, and intercept = 0.04 for 90 mmHg; r = 0.98, slope = 0.94, and intercept = 0.09 for 70 mmHg; r = 0.98, slope = 0.96, and intercept = 0.07 for 50 mmHg). Blood flow values (ml x min-1 x 100 g-1) of right and left bone samples determined at the different arterial blood pressures were identical. During moderate hypotension (70 mmHg MAP), blood flow in all bone samples remained unchanged compared with 90 mmHg MAP, whereas a significant decrease of bone blood flow was observed at severe hypotension (50 mmHg MAP). Our results demonstrate that the FM technique is valid for measuring bone blood flow. Differences in bone blood flow during altered hemodynamic conditions can be detected reliably. In addition, changes in bone blood flow during hypotension indicate that vasomotor control mechanisms, as well as cardiac output, play a role in setting bone blood flow.

Age-related changes of the glenoid labrum in normal shoulders.

A growing number of labral changes are described in the literature. The purpose of this study was to evaluate the glenoid and labrum of normal shoulders at different ages and characterize any apparent age-dependent changes. We analyzed 32 normal cadaveric shoulders with a mean age of 57 years (range, 18-89 years). There were 22 male and 10 female cadavers, with 14 right and 18 left specimens. The shoulders were studied macroscopically, histologically, and radiologically. The radiologic evaluation consisted of an analysis of the subchondral mineralization of the glenoid with the use of computed tomographic osteoabsorptiometry. Macroscopically, there were no statistically significant differences among the age groups. Histopathologically, the labrum showed a significant qualitative and quantitative increase (P <.01) in lesions across all regions with increasing age. In younger individuals, lesions at the 12-o'clock position were the most prevalent, with the incidence increasing with age. The anterosuperior position was the region with the next highest prevalence. This was also the area of the highest stress distribution on the glenoid. Our studies demonstrated clear histopathologic changes of the glenoid labrum that are significantly age-related at specific sites. The earliest changes are seen close to the area of highest stress distribution of the glenoid, which could explain the progressive labral changes with increasing age. Arthroscopically detected changes of the glenoid labrum should be evaluated in the context of age-related changes in normal shoulders.

Long-term results of high tibial osteotomy for medial osteoarthritis of the knee.

We reviewed retrospectively the results in patients who had undergone one hundred and four high tibial lateral osteotomies. The operations were all performed between 1985 and 1993. Each one of fifty men and forty nine women demonstrated a varus deformity of the knee with a coexistent medial osteoarthritis. Results were reviewed in 49 patients (62 knees) with an average follow-up of 10.2 years (range 6-14 years). Of the remaining 42 patients, 8 were lost to follow-up, 10 had died, and 24 were subsequently treated with total knee arthroplasty at an average 4.7 years after having had a high tibial osteotomy. Clinical results were evaluated using the Hospital for Special Surgery Score (HSS) and the Knee Society Score. Radiographs were systematically analysed to evaluate osteoarthritis and leg axis. Forty four (90 per cent) of the forty nine patients stated the results met their expectations and given the same circumstances, they would have the operation once again. In these patients the knee score results were excellent. The same patients had excellent HSS and Knee Society Scores. Five patients (10 per cent) had a poor result and twenty four patients were treated later by total knee arthroplasty because of pain. The following factors set these patients apart from those with more favorable results: previous arthroscopic debridement, obesity, lateral knee osteoarthritis, insufficient valgus correction, and an age of more than 55 years. High tibial valgus osteotomy provides good pain relief and improved function in carefully selected patients. Our results support this conclusion.

[Open surgical procedures in calcifying tendinitis of the shoulder - concomitant pathologies affect clinical outcome]. / Offene Operationsverfahren bei Tendinosis calcarea der Schulter - Begleitpathologien beeinflussen das Operationsergebnis.

AIM OF THE STUDY: Open surgical removal of calcifications in chronic courses of calcifying tendinitis of the shoulder can be combined with acromioplasty. Independent of the surgical procedure not all patients achieve satisfactory surgical results. The aim of the study was to investigate whether preoperatively known epidemiologic, social, clinical and radiologic factors or intraoperative findings might influence the therapeutic outcome. METHODS: Following diagnostic arthroscopy, open removal of the calcifications was done as an isolated procedure (group A, n = 12) or combined with open acromioplasty (group/B, n = 24). Follow-up was 33 months for both groups. RESULTS: Clinical outcomes were comparable in both groups (Group A, 74.9 points; Group B 73.4 points, Constant-Murley score) and independent of gender, age, profession, duration of anamnesis, hospital-stay period, follow-up period, dominance of arm, preoperative Constant-Murley score, calcification morphology and size and acromial type. 20 patients in total achieved a clinical outcome of

[The sublabral foramen. Congenital or an acquired variance?]. / Das Foramen sublabrale. Eine angeborene oder eine erworbene Variante?

The sublabral foramen is considered to be a clinically asymptomatic, isolated variant of the anterior-superior capsulo-labral-complex. It may be observed during shoulder arthroscopy and may implicate problems in differential diagnosis of traumatic lesions. In an anatomic study on 89 macroscopically healthy shoulder specimens the age distribution and the incidence in relation to the varying shape of the glenoid as feasible factors of influence for a sublabral foramen (SF) were analyzed in order to elucidate the unknown pathogenesis of SF. In addition histologic sections of three SF samples were obtained. Mean age of the specimens with a sublabral foramen [n=20; 69 (37-84) years] was significantly higher (p=0.04) compared to samples without a sublabral foramen [n=69; 59 (18-94) years]. An increased incidence of a sublabral foramen in relation to a distinct shape of the glenoid could not be established although a prevalence of a glenoid with anterior notch was observed. The results indicate an age-related development of the sublabral foramen,thus in younger patients with an anterior-superior capsulolabral displacement local signs of trauma and involvement of the biceps anchor should be controlled before definitive diagnosis.

Subchondral mineralization patterns of the glenoid after tear of the supraspinatus.

Clinical experience has shown that in many cases of rupture of the supraspinatus tendon the head of the humerus shifts upward against the acromion. This implicates alterations in the stress acting on the joint surface. Accordingly, this should be reflected by changes in the distribution of the subchondral mineralization on the glenoid. Computed tomography osteoabsorptiometry was used to evaluate the distribution patterns on the subchondral bone plate of the glenoid in 67 human shoulders. The position of the two most frequent density maxima was determined by a standard procedure. In macroscopically normal specimens (n = 24) every case had an anterosuperior density maximum and the majority (75%) also had a posterior positioned maximum. In some instances a third maximum, placed centrally (17%) or in an anteroinferior position (4%) was seen. In shoulders with a torn supraspinatus tendon (n = 43) a third density maximum frequently appeared in the center of the glenoid (42%); however, most notable was a significant change of the posterior maximum position toward central and superior. With increasing size of the tendon defect an extension of the shift of the posterior maximum (superiorly and centrally) was observed in contrast to the constant anterior maximum. The changes in the distribution of glenoid subchondral bone mineralization in shoulders with a supraspinatus tear must be regarded as a morphologic parameter for the altered long-term stress acting on the joint surface.