Correction: Winkler et al. The Adhesion G-Protein-Coupled Receptor GPR115/ADGRF4 Regulates Epidermal Differentiation and Associates with Cytoskeletal KRT1. Cells 2022, 11, 3151

In the original publication [...].

Does the intraoperative 3D-flat panel control of the planned implant position lead to an optimization and increased in safety in the anatomically demanding region C1/2?

BACKGROUND

The aim of this study was to evaluate the applicability and advantages of intraoperative imaging using a 3D flat panel in the treatment of C1/2 instabilities.

MATERIALS

Prospective single-centered study including surgeries at the upper cervical spine between 06/2016 and 12/2018. Intraoperatively thin K-wires were placed under 2D fluoroscopic control. Then an intraoperative 3D-scan was carried out. The image quality was assessed based on a numeric analogue scale (NAS) from 0 to 10 (0 = worst quality, 10 = perfect quality) and the time for the 3D-scan was measured. Additionally, the wire positions were evaluated regarding malpositions.

RESULTS

A total of 58 patients were included (33f, 25 m, average age 75.2 years, r.:18-95) with pathologies of C2: 45 type II fractures according to Anderson/D'Alonzo with or without arthrosis of C1/2, 2 Unhappy triad of C1/2 (Odontoid fracture Type II, anterior or posterior C1 arch-fracture, Arthrosis C1/2) 4 pathological fractures, 3 pseudarthroses, 3 instabilities of C1/2 because of rheumatoid arthritis, 1 C2 arch fracture). 36 patients were treated from anterior [29 AOTAF (combined anterior odontoid and transarticular C1/2 screw fixation), 6 lag screws, 1 cement augmented lag screw] and 22 patients from posterior (regarding to Goel/Harms). The median image quality was 8.2 (r.: 6-10). In 41 patients (70.7%) the image quality was 8 or higher and in none of the patients below 6. All of those 17 patients the image quality below 8 (NAS 7 = 16; 27.6%, NAS 6 = 1, 1.7%), had dental implants. A total of 148 wires were analyzed. 133 (89.9%) showed a correct positioning. In the other 15 (10.1%) cases a repositioning had to be done (n = 8; 5.4%) or it had to be drawn back (n = 7; 4.7%). A repositioning was possible in all cases. The implementation of an intraoperative 3D-Scan took an average of 267 s (r.: 232-310 s). No technical problems occurred.

CONCLUSION

Intraoperative 3D imaging in the upper cervical spine is fast and easy to perform with sufficient image quality in all patients. Potential malposition of the primary screw canal can be detected by initial wire positioning before the Scan. The intraoperative correction was possible in all patients. Trial registration German Trials Register (Registered 10 August 2021, DRKS00026644-Trial registration: German Trials Register (Registered 10 August 2021, DRKS00026644- https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00026644 ).

[Spinal deformity in neurofibromatosis type 1]

BACKGROUND

With a prevalence of up to 60%, spinal deformity represents the most common skeletal manifestation of neurofibromatosis type 1. The deformity can occur as a non-dystrophic or as a less common dystrophic type. This distinction is of great relevance because the therapeutic strategy is completely different in each case.

NON-DYSTROPHIC TYPE

The non-dystrophic type can be treated like idiopathic scoliosis due to the comparable behavior of both entities. However, care must be taken regarding the so-called modulation. Modulation describes the formation of dysplasias of the spine. This will result in a progression behavior as known from the dystrophic type.

DYSTROPHIC TYPE

For the dystrophic type, different spinal dysplastic changes are typical. These lead to a rapid progression of deformity and a lack of response to conservative treatment. If untreated, severe and grotesque deformities can arise. This type of deformity requires early surgical intervention, even in childhood. The knowledge about the peculiarities of this disease in general, as well as the typical changes of the spine are prerequisites to managing these often-challenging situations.

Spontaneous intraoperative lumbar fracture leading to an unexpected correction in ankylosing spondylitis corrective surgery - a case report

Severe kyphotic deformity in patients with ankylosing spondylitis can be corrected surgically to achieve a better spinal alignment and an improved visual axis. Different surgical techniques are used today depending on the extent of ossification and the degree of kyphosis. It is well known that the underlying disease leads to distinct biomechanical changes of the spinal column causing an increased fracture risk especially in case of minor trauma. This includes manipulations during surgical procedures as well as during the required perioperative measures.

We present the case of a 45-year-old patient with severe global kyphotic deformity due to ankylosing spondylitis. During the elective corrective surgery (pedicle subtraction osteotomy at the level of L3) the patient sustained a spontaneous fracture at L2/3. This fortunately nondisplaced wedge-shaped fracture in the sense of a Smith-Peterson osteotomy led to a spontaneous correction of the kyphosis. The described unexpected event required a change in the surgical strategy. Correction could be achieved using a two-stage surgical procedure without further drawbacks for the patient.

This case report stresses the need of particular attention regarding the increased susceptibility of the spinal column in case of ankylosing spondylitis.

Which anatomic structures are responsible for the reduction loss after hybrid stabilization of osteoporotic fractures of the thoracolumbar spine?

Introduction

Hybrid stabilization is an accepted therapy strategy for unstable osteoporotic thoracolumbar fractures. However, a moderate reduction loss has been reported and it remains unclear which anatomic structure is responsible for the reduction loss.

Methods

This retrospective study was performed at a level I trauma center. Patients aged 61 and older were stabilized using hybrid stabilization after suffering acute and unstable osteoporotic vertebral body fractures at the thoracolumbar spine. Posterior stabilization was done short-segmental and minimal invasive with cement-augmentation of all pedicle screws. The minimum follow-up has been 2 years. The outcome parameters were the reduction loss and the relative loss of height of both intervertebral discs adjacent to the fractured vertebral body, the fractured vertebral body and a reference disc (intervertebral disc superior of the stabilization) between the postoperative and latest lateral radiographs. Additionally, implant positioning and loosening was analyzed.

Results

29 mainly female (72%) patients (73.3 ± 6.0 years) were included. Fractures consisted of 26 incomplete burst fractures and 3 complete burst fractures of the thoracolumbar junction (Th11 – L2: 86%) and the midlumbar spine. The mean follow-up time was 36 months (range: 24–58 months). The mean reduction loss was 7.7° (range: 1–25). The relative loss of heights of both intervertebral discs adjacent to the fractured vertebral body, the reference disc, and the central vertebral body were significant. Thereby, the relative loss of the superior disc height was significant higher compared to the reference disc. Additionally, only the relative loss of central vertebral body height and reduction loss correlated significantly. There were no signs of implant loosening in any patient.

Conclusions

The mean reduction loss was moderate 3 years after hybrid stabilization of unstable osteoporotic vertebral fractures of the thoracolumbar spine. A significant loss of both adjacent disc heights and the central vertebral body was seen, with the highest loss in the superior adjacent disc significantly outranging the reference disc. The superior adjacent intervertebral disc and the central part of the fractured vertebral body seem to be responsible for the majority of reduction loss.

Halo Fixator and Halo Traction - Value for the Treatment of Spinal Disorders in Childhood

The application of the Halo fixateur in case of spinal pathologies in childhood is a standardized technique. The halo fixateur may be used for treatment of injuries of the cervical spine, for additional stabilization following extended surgery at the cervical spine and their transitional regions as well as to achieve preoperative reduction in case of severe and rigid deformity. These indications are, referred to the early age, rare. However, the successful use of the Halo fixateur presumes a certain familiarity with the device and experiences regarding the underlying diseases to minimize related risks and to avoid possible complications. In this article the use and specific features regarding the application of the halo fixateur in childhood based on presented cases and the literature will be discussed.

Odontoid fractures in combination with C1 fractures in the elderly treated by combined anterior odontoid and transarticular C1/2 screw fixation

INTRODUCTION

The purpose of this study was to evaluate risk factors of accompanied C1 fractures in elderly patients with type II odontoid fractures (OF) and to analyze the complication rate and the outcomes of patients after combined anterior odontoid and transarticular C1/2 screw fixation (AOTAF).

MATERIALS AND METHODS

The study represents a retrospective case series at a single level-1 trauma center. All elderly patients (≥ 70 years) with acute combination injuries (CI) including type II OF with an accompanied C1 fracture, who were treated by an anterior approach, were included. All postoperative complications were analyzed based on the patient notes. Clinical and radiological controls were performed after 1 year. Main parameters of interest were 1-year mortality rate, pain level, and satisfaction rate after 1 year.

RESULTS

A total of 23 patients were included. The average age was 84.6 years (range 73-94 years). All patients had atlanto-odontoid osteoarthritis (AOO) and all but two patients were injured by low-energy falls. Dysphagia was the most common postoperative complication (26.1%). Surgical revision was necessary in one of these patients due to hematoma. Dysphagia improved in all patients considerably. Loss of follow-up was 21.7%. The 1-year mortality was 21.7% (n = 5). The mean pain level and satisfaction rate was 2.5 (± 0.9) and 7.3 (± 0.7), respectively. After 1 year, no signs of non-union were visible.

CONCLUSIONS

AOO was observed in all patients with CI. The main cause of trauma was a low-energy fall. The pain levels were low to moderate and satisfaction levels were promising 1 year after surgery. Nonetheless, AOTAF is associated with a high rate of postoperative dysphagia, which resolves in the majority of patients due to conservative management.

[Release of moveable segments after dorsal stabilization : Impact on affected discs]

BACKGROUND

Bisegmental dorsal stabilization is a common treatment option for instable compression fractures of the thoracolumbar spine; however, it remains unknown to what extent bridging compromises intervertebral discs.

OBJECTIVES

The purpose of this study was to determine the disc height and functional features in comparison to healthy intervertebral discs after removal of the dorsal fixator and particularly under consideration of the time span between dorsal stabilization and implant removal (IR).

MATERIAL AND METHODS

The IR was performed in 19 patients after an average of 13 months (range 8-24 months) after dorsal stabilization of instable vertebral compression fractures of the thoracolumbar junction and lumbar spine. An additional ventral monosegmental spondylodesis was performed in 10 patients with incomplete burst fractures. Thus, a total of 28 intervertebral discs were temporarily bridged (bridged discs), with an adjacent endplate fracture in 10 (injured discs) and no adjacent bony lesion in 18 discs (healthy discs). The intervertebral discs superior and inferior to the instrumentation were selected as controls (control discs). Standardized conventional lateral radiographs were taken prior to and after IR as well as after 6 months. Additionally, standardized lateral radiographs in flexion and extension were taken. The intervertebral disc height (disc height) was determined by two independent board approved orthopedic observers by measuring the anterior, central and dorsal intervertebral disc spaces on all lateral radiographs as well as the intervertebral disc angles (disc angle) defined by the intervertebral upper and lower endplates in the flexion and extension views. Intradisc function (disc function) was defined as the difference between the disc angle in extension and flexion. The measurements were repeated after 12 months. Univariate analysis was performed using ANOVA and significance was set at p < 0.05. Interobserver and intraobserver comparisons of the disc heights and the disc angles were determined with intraclass correlation coefficients.

RESULTS

No significant differences were seen in disc function and disc height between the controls and the bridged discs at all times of measurement; however, injured discs showed a significantly reduced disc height and disc angle in extension compared to healthy discs (p = 0.028 and p = 0.027, respectively). Additionally, patients with IR during the first 12 months had significantly reduced disc heights compared to those patients with delayed IR within the second postoperative year (p = 0.018). Interobserver and intraobserver agreement for disc function was 0.80 (95 % confidence interval CI: 0.68-0.88) and 0.85 (95 % CI 0.76-0.90), respectively. The interobserver and intraobserver correlations for disc height were 0.85 (95 % CI: 0.76-0.90) and 0.93 (95 % CI 0.88-0.95), respectively.

CONCLUSION

Bridging of an intervertebral disc with IR within 24 months does not cause immediate loss of disc function or reduction of disc height; however, temporary bridging in combination with an adjacent endplate fracture causes significant reduction of disc height and loss of extension. Additionally, no beneficial effects could be seen by reducing the time span between stabilization and IR to below 12 months.

Skeletal muscle expression of the adhesion-GPCR CD97: CD97 deletion induces an abnormal structure of the sarcoplasmatic reticulum but does not impair skeletal muscle function

CD97 is a widely expressed adhesion class G-protein-coupled receptor (aGPCR). Here, we investigated the presence of CD97 in normal and malignant human skeletal muscle as well as the ultrastructural and functional consequences of CD97 deficiency in mice. In normal human skeletal muscle, CD97 was expressed at the peripheral sarcolemma of all myofibers, as revealed by immunostaining of tissue sections and surface labeling of single myocytes using flow cytometry. In muscle cross-sections, an intracellular polygonal, honeycomb-like CD97-staining pattern, typical for molecules located in the T-tubule or sarcoplasmatic reticulum (SR), was additionally found. CD97 co-localized with SR Ca²⁺-ATPase (SERCA), a constituent of the longitudinal SR, but not with the receptors for dihydropyridine (DHPR) or ryanodine (RYR), located in the T-tubule and terminal SR, respectively. Intracellular expression of CD97 was higher in slow-twitch compared to most fast-twitch myofibers. In rhabdomyosarcomas, CD97 was strongly upregulated and in part more N-glycosylated compared to normal skeletal muscle. All tumors were strongly CD97-positive, independent of the underlying histological subtype, suggesting high sensitivity of CD97 for this tumor. Ultrastructural analysis of murine skeletal myofibers confirmed the location of CD97 in the SR. CD97 knock-out mice had a dilated SR, resulting in a partial increase in triad diameter yet not affecting the T-tubule, sarcomeric, and mitochondrial structure. Despite these obvious ultrastructural changes, intracellular Ca²⁺ release from single myofibers, force generation and fatigability of isolated soleus muscles, and wheel-running capacity of mice were not affected by the lack of CD97. We conclude that CD97 is located in the SR and at the peripheral sarcolemma of human and murine skeletal muscle, where its absence affects the structure of the SR without impairing skeletal muscle function.

3D-based navigation in posterior stabilisations of the cervical and thoracic spine: problems and benefits. Results of 451 screws

INTRODUCTION

Navigated procedures in spinal surgery have been established due to an increasing demand for precision. Especially, 3D C-arms connected to navigation systems are being used more often and can be utilised intraoperatively for the planning and controlling of screw positions. This prospective study analyses our experiences with 3D-based navigation in posterior stabilisations in the cervical and thoracic spine.

METHODS

A 3D C-Arm (Ziehm Vision Vario 3D(®)) was connected to a navigation system (VectorVision, Brainlab(®)) and used for the placement of, in total, 451 screws among 67 patients. Of those, 14 patients had to undergo operations in the cervical and 53 in the thoracic spine. Postoperatively, the positioning was observed with computed tomography (CT).

RESULTS

The application time is approximately 6 min. In total, 354/451 (78.5%) screws could be inserted assisted with navigation, and 272/451 (60.3%) were controlled intraoperatively. Regarding the cervical spine, in 87.1% (61/70) of the screws, the navigation procedure was uneventful. The positioning of 63.2% (43/68) of the screws was checked intraoperatively. In the upper thoracic spine, 77% (293/381) could be placed with navigation and 59.6% (227/381) were controlled intraoperatively. Occasionally, the scanning setup was problematic. Correct placement was seen in 92.7% of screws; for the remaining screws, no revision was needed.

CONCLUSIONS

Intraoperative 3D imaging navigation for posterior spinal stabilisations is technically feasible and reliable in clinical use. The image quality depends on the individual bone density. With undisturbed visibility of the vertebral body, the reliability of 3D-based navigation is comparable to that of CT-based procedures. Additionally, it has the advantage of skipping the preoperative acquisition of data as well as the matching process, with reduced radiation doses.

Revision characteristics of cement-augmented, cannulatedfenestrated pedicle screws in the osteoporotic vertebral body: a biomechanical in vitro investigation

In generalized osteoporosis, instrumentation with cement-augmented pedicle screws is an amplification of the therapeutic spectrum. Early clinical results are promising for both solid and cannulated screws; however, there are concerns regarding the revision characteristics of these screws, especially for the cannulated-fenestrated type with its continuous cement interconnection from the core of the screw to surrounding bone tissue. In a human cadaver model, bone mineral density (BMD) was assessed radiographically. Spinal levels T9–L4 were instrumented left unilaterally, transpedicularly by using cannulated-fenestrated pedicle screws with the dimensions 6.5 × 45 mm. Polymethylmethacrylate cement (1.5 ml) was injected through the screws into each vertebra. After polymerization of the cement, the extraction torque was recorded. For both implantation and explantation of the screws, a fluoroscope was used to guarantee correct screw and cement positioning and to observe possible co-movements—that is, any movement of the cement mass within the vertebral body upon removal of the screw. For comparison, the extraction torque of same-dimension pedicle screws was recorded in a nonosteoporotic, non–cement-augmented instrumentation. The BMD was 0.60 g/cm2, a level that corresponds to a severe grade of osteoporosis. For removal of the screws, the median and mean extraction torques were 34 and 49 ± 44 Ncm, respectively. No co-movements of the cement mass occurred within the vertebral body. In the nonosteoporotic control, BMD was 1.38 g/cm2. The median and mean extraction torques were 123 and 124 ± 12 Ncm, respectively. Thus, the revision characteristics of cement-augmented, cannulated-fenestrated pedicle screws are not problematic, even in cases of severe osteoporosis. The winglike cement interconnection between the screw core and surrounding bone tissue is fragile enough to break off in the event of an extraction torque and to release the screw. There is no proof to support the theoretical fear that while trying to remove a screw, the composite of screw and cement would not break but instead would rotate as a whole in the osteoporotic vertebral body.

[Quality of scientific advice to politics. Lecture at the Berlin-Brandenburg Academy of Science and Humanities]

Scientific advice to politics is a primary function of governmental research. The advisory process is, in the ideal situation, a collective duty of science and politics. The final decision rests ultimately with politicians. An understanding of the differences between science and politics is necessary for successfully providing advice to politicians. The requirements necessary to allow politics to substantially follow the advice of scientists are multifarious. The first of these is trust from the side of politics and the public and from the side of science competitive research, respect and communication skills, neutrality and integrity. From these requirements it is possible to derive criteria for quality assurance in advice to politics. The maintenance of scientific expertise at the competitive international level demands independent, qualified and adequately financed research. Governmental institutes have an antenna function: they have to recognize in good time whether risks are increasing, whether the government has to be informed and whether there is a need for action. The continuing maintenance of excellence requires measures of quality assurance at all levels. Evidence for the quality of advice to politics can, for example, be found in the good reputation of an institution and its prominent representatives. Success in research is an indirect quality criterion that can be and should be measured to a certain extent. The influence of advisory activities on political decisions is direct evidence for the quality of the advice. A classic example of highly successful policy advice is the development of the German AIDS policy.

[Differentiated therapy in necrotizing fasciitis of three extremities]

AIM

Here we present the clinical symptomatology of and therapy for necrotizing fasciitis.

METHOD

The case of a 35-year-old female patient with sustaining fractures of the 5 (th) and 10 (th) thoracic vertebrae and a pulmonary contusion and without any skin lesions is presented.

RESULTS

Conventional x-rays and computed tomography revealed stable spine fractures not necessitating surgical intervention. Fifteen days after the accident the patient developed septic conditions. An interdisciplinary search (surgical, neurological, urological, internal medicine) for the septic focus first remained negative. After demarcation of necrotic skin areas at the upper left arm, bilateral necrotizing fasciitis was diagnosed at both thighs and at the lower left leg, necessitating continuous optimisation of the therapeutic strategy.

CONCLUSION

Local aggressive surgical therapy in combination with systemic antibiotic administration is the therapy of choice in treatment of the necrotizing fasciitis. It should be performed according to the principle "life before limb". In the presented case the patient recovered and good functional results could be achieved.

The endoscopically assisted simultaneous posteroanterior reconstruction of the thoracolumbar spine in prone position

BACKGROUND CONTEXT

The reconstruction of the anterior column of the thoracolumbar spine has become more common in the last few years, due largely to the unfavorable results of exclusively posterior surgical treatment, which has been associated with a lack of about 10 degrees of kyphosis correction after removal of the instrumentation. The minimally invasive anterior techniques have reduced the morbidity of the anterior approach significantly.

PURPOSE

A minimally invasive technique for anterior stabilization of the spine may reduce the morbidity of the open approach. Irrespective of an anterior open or an endoscopic approach, the posteroanterior instrumentation of thoracolumbar fractures requires time-consuming intraoperative maneuvers to change the patient position from prone to lateral. We describe here a standardized anterior endoscopically assisted approach for the segments T4 to L4. This approach allows the patient to remain in prone position. A 4- to 5-cm incision combined with a retractor system is used.

STUDY DESIGN/SETTING

In a prospective study, all patients of our clinic who underwent surgery of the thoracolumbar spine between July 1999 and May 2001 were registered. Study criteria were duration of surgery, duration of anesthesia, intra- and postoperative complications.

PATIENT SAMPLE

Between July 1999 and May 2001, 42 patients (25 male, 17 female, average age of 41.9 years), who presented with 55 injured spinal levels and underwent surgery of the thoracolumbar spine in prone position, were included.

OUTCOME MEASURES

Duration of surgery (posterior/anterior/total), duration of anesthesia, method of instrumentation, intra- and postoperative complications, postoperative hospital stay and radiographs were evaluated.

METHODS

Surgery was performed in prone position. A thoracic approach was used for instrumentation of T9 to L2. A retroperitoneal approach was used for stabilization of L1 to L5. Both procedures were endoscopically assisted with a new retractor system (Synframe; Synthes GmbH, Umkirch, Germany). In this manner, only an incision 4 to 5 cm long and a stab incision for the endoscope were required. The whole procedure was performed in prone position without a change of position during surgery.

RESULTS

A total of 42 patients underwent surgery following this technique: 14 isolated anterior procedures (median duration of surgery, 181 minutes); 13 simultaneous one-stage procedures (median duration of surgery: 210 minutes) and 15 combined two-stage procedures (median duration of surgery: 90 minutes posterior, 120 minutes anterior, 240 minutes posterior+anterior). In the simultaneous posteroanterior procedures, the anterior instrumentation was performed 20 times using one rod, twice using two rods and in six patients simply by bone grafting. No intraoperative complications were observed. In the postoperative course, one case of pneumothorax, one case of hemothorax and one case of transient intercostal neuralgia occurred.

CONCLUSION

The approach to the anterior spine in prone position is feasible by using a self-holding retractor system for the region between T4 and L4. The duration of anesthesia for the one-stage simultaneous procedure was reduced by about 40 minutes, because changing the position of the patient is no longer necessary. The minimal incision, in combination with the retractor system, significantly reduces cost by allowing the use of less expensive instruments and implants. The advantages of the open and the endoscopic techniques are combined, while their disadvantages are minimized. The main advantage of the prone position is the opportunity to access the anterior and posterior spine simultaneously, which is especially helpful in reduction maneuvers.

[Thoracolumbar spine fractures after conservative and surgical treatment. Dependence of correction loss on fracture level]

QUESTION

This retrospective study presents results after conservative and operative treatment of thoracolumbar fractures as function of its localization.

METHODOLOGY

In 2 years 70 patients with A1/A2 fracture were conservatively treated, 38 patients with A3/B/C injury were treated by internal fixtor. For evaluation 3 vertebral sections(Th5-10,Th11-L2,L3-5)were defined. Follow-up took place 1 year after implant removal or end of conservative treatment.

RESULTS

The correction-loss was highest in thoracic, lowest in lumbar region. After conservative therapy,correction-loss was located to 3/4 in vertebra itself, after operative treatment especially in adjacent disc spaces. There was no general correlation to complaints.

CONCLUSION

In consequence of these results A1/A2-fractures in the upper thoracic spine (<Th10) with kyphosis >15 degrees will be stabilized anteriorly, in other regions functional treated. A3-fractures of thoracic spine and thoracolumbar junction will be operated from anterior, in lower lumbar spine (>L3) from dorsal. B- and C-injuries should be instrumented with a combined dorsoventral procedure.

Arthroscopic-assisted simultaneous reconstruction of the posterior cruciate ligament and the lateral collateral ligament using hamstrings and absorbable screws

Arthroscopic-assisted simultaneous reconstruction of the posterior cruciate ligament (PCL) and the lateral collateral ligament (LCL) using hamstring tendon grafts is described. The femoral tunnel is drilled through an incision over the medial femoral condyle and the tibial tunnel through the same skin incision used for harvesting the tendon graft. PCL reconstruction is performed using a 4-strand hamstring tendon graft and absorbable screw fixation. The tendon of the semitendinosus muscle of the uninvolved knee is used as a lateral loop for LCL reconstruction. After pulling the transplant through the fibular head, femoral fixation of the loop is made with an absorbable screw.

[Crossed screw osteosynthesis of proximal humerus fractures]

Between March 1997 and October 1999 thirty-one patients with displaced proximal humeral fractures were treated with crossed screw osteosynthesis. Insertion of the screws was realized by using a deltoideo-pectoral approach placing the screws anteriorly and posteriorly in a crossed manner from the distal fragment into the humeral head. Additionally, in all two-part-fractures a tension band was applied. In all three-part-fractures, the greater tuberosity was reattached by additional screws. In 21 patients (14 female, 7 male, median age 62 years, 18-86) a clinical and radiological follow-up (median 18 months, 10-29) was obtained. Fractures were classified as two-part-fractures in 10 patients and as three-part-fractures in 11 patients. According to the Constant-Score, "excellent" and "good" results were achieved in 15 patients, "moderate" results were found in 3 patients. However, in 3 patients results were only "poor" (1 two-part-, 2 three-part-fractures). The complication rate was 29% (premature hardware removal due to head perforation in 3 cases; humeral head necrosis necessitating prosthetic replacement in 2 patients; secondary displacement in 1 case). Crossed screw osteosynthesis represents an justified alternative in the surgical treatment of displaced proximal humeral fractures permitting early functional therapy.