Objective To summarize the efficacy of using proximal humeral intramedullary pins in the treatment of two-part surgical neck of the proximal humerus and the points to be noted during and after surgery through a case follow-up study. Methods Twenty-two patients with fresh fractures of the two-part surgical neck treated with proximal humeral intramedullary pins were followed up for a mean of 13 months. The average age of the patients was 57 years. Final follow-up was performed by taking shoulder x-rays to evaluate healing and applying the VAS score, ASES score, Constant-Murley score, UCLA score, and SST questionnaire assessment. Results All 22 patients had initial healing of their fractures within 8 weeks postoperatively. There were no infections, humeral head necrosis, or any complications related to the internal fixation during the follow-up period. The patients had a mean postoperative active forward flexion supination of 147.8°, a mean active lateral external rotation of 45.5°, and a mean active internal rotation of T10. The mean VAS score for postoperative pain was 1.5, the mean ASES score was 81.2, Constant-Murley score was 85.4, UCLA score was 29.9, and SST score was 9.5. 18 patients had excellent shoulder function, but 4 patients had poor shoulder function. All patients were satisfied with the surgical results. Conclusion Under the premise of strict selection of surgical indications, careful intraoperative operation according to the characteristics of the fracture and internal fixation, and good postoperative rehabilitation guarantee, closed reduction proximal humeral intramedullary pin (PHN) fixation is an effective surgical modality for the treatment of proximal humeral fractures. Zhu Yiming, Department of Sports Injury, Beijing Jishuitan Hospital? IntroductionProximal humeral fractures are a common type of fracture in clinical practice and are reported to account for about 4-5% of all fractures. Its occurrence is often closely related to the gradual onset of osteoporosis with age, which can be seen in its prevalence in the elderly, especially in older women, who are significantly more likely to suffer from it than younger people. Most proximal humerus fractures are nondisplaced and are better treated conservatively. However, for those fractures with significant displacement, conservative treatment may not be satisfactory. Due to its own characteristics, the proximal humerus fracture has been one of the more difficult fractures to manage in clinical practice. In this study, we report our clinical experience with the application of the proximal humeral intramedullary pin (PHN, Synthes) in the treatment of proximal humeral fractures, and discuss the indications for the application of this internal fixation and the problems that may be encountered intraoperatively and postoperatively. Materials and Methods A total of 24 patients suffering from proximal humeral fractures were treated with the proximal humeral intramedullary pin (PHN) in our department from March 2005 to July 2006, of whom 2 were lost to follow-up, so the remaining 22 cases were included in the study. The mean age of the patients was 57 years (25-78), of which 7 were male and 15 were female. 7 patients had primary lateral involvement and the remaining 15 had non-primary lateral involvement. The diagnosis in all patients was a two-part surgical neck fracture of the proximal humerus, which was fresh in all cases and operated on for a mean of 10 days (4 -23 days) from the time of injury. The surgery was performed with closed reduction and PHN intramedullary pin fixation in all cases. Patients were reviewed at 3 weeks, 6 weeks, 3 months, 6 months, and 1 year after surgery; thereafter, they were reviewed annually. Orthopantomographies of the scapula in the rotated neutral position, as well as lateral and axillary radiographs of the scapula were taken at each review to check the fracture healing and the position of the internal fixation. The ASES shoulder function scale, UCLA scale, Constant-Murley scale and SST scale were completed by the physician at each annual review six months after surgery and thereafter to assess the patient’s shoulder function. The VAS score (0-10, 0 being no pain) was also recorded to assess the patient’s shoulder pain. The patients’ shoulder function was graded according to their UCLA scores, with >34 being excellent, 29-33 being good, and ≤28 being poor. The unique design of the Proximal Humeral Nail (PHN) introduced by Synthes for proximal humeral fractures is that the proximal locking nail is designed with a spiral blade as opposed to the previous screw-like proximal locking nail, which allows for a larger gap between the proximal humerus and the more lax proximal bone. This increases the contact area with the more lax proximal humeral bone, thus reducing the stress at the bone-internal fixator contact interface. The spiral blade is squeezed by the end cap of the intramedullary pin into the hole in the proximal end of the main nail that is used to pass through the spiral blade, thus achieving angular stabilization. The main intramedullary pin is 7.5 mm in diameter and 150 mm long, and the proximal locking nail has an optional 3.9 mm diameter diagonal locking nail in addition to the rotating blade, while the distal end is locked with two 3.9 mm diameter locking nails. Surgical approach All patients were anesthetized intraoperatively with interosseous sulcus. After successful anesthesia, the patient was placed in the supine position and the patient’s head was properly secured. The G-arm or C-arm for fluoroscopy was properly positioned before surgery to ensure that orthogonal and axillary fluoroscopic images of the affected shoulder could be obtained intraoperatively. The incision is located on the anterolateral aspect of the patient’s shoulder crest and is made in the direction of the local Langer’s line, anteriorly and posteriorly. The incision is usually about 3-4 cm, and the subcutaneous tissues are fully freed after the incision to reveal the anterior angle of the acromion. From the anterior angle of the acromion, the deltoid muscle is split downward along the anterior and middle deltoid gaps to the level of about 4 cm distal to the anterior angle of the acromion. The subacromial bursa and the subdeltoid bursa are fully separated to reveal the humeral head and the rotator cuff tissue overlying it, and the inter-nodal sulcus is palpated anteriorly. The supraspinatus tendon is incised approximately 1.5 cm posteriorly along the course of the supraspinatus tendon, and the incision is approximately 1 cm long. Under fluoroscopic guidance, the medial humeral tuberosity is opened at the depression of the junction between the medial aspect of the greater tuberosity and the outer edge of the humeral head joint surface. After fluoroscopic closure and repositioning of the fracture, the PHN intramedullary needle is inserted from the open medullary site into the distal medullary cavity. After confirming the insertion of the intramedullary pin in place under fluoroscopy, the proximal rotating blade and distal locking nail were inserted by relying on the sight on the intramedullary pin holder to determine the position of the proximal rotating blade and distal locking nail. Fluoroscopic examination of the fracture is performed, and if possible, proximal oblique locking nails are inserted to enhance the reliability of fixation. The end cap of the intramedullary pin is screwed in from the end of the pin and the rotating fin is compressed. The wound is flushed and a drainage tube is left in place. With a #2 Ethibond wire, the incision over the supraspinatus tendon is closed, taking care not to leave the knot above the supraspinatus tendon to release later signs of impingement. Finally, the deltoid is firmly reconstructed and the wound is closed. The affected limb is protected by a postoperative neck and wrist sling. The patient began passive joint exercises under the guidance of a physical therapist the day after surgery. At 6 weeks postoperative review, the sling was removed and active joint exercises were performed when the X-ray showed scab growth. At 3 months postoperatively, the affected limb was able to perform daily activities basically without restriction. Statistical analysis: We applied the paired t-test in the spss11.5 software package to test whether the differences between the supraspinatus muscle strength and the mean muscle strength of the external rotation muscle groups (infraspinatus and teres minor) on the affected side and the healthy side were significant. In addition, we applied the multiple linear regression analysis in this package to elucidate whether there was a correlation between the patient’s preoperative and postoperative factors and the functional recovery of the shoulder joint. The mean operative time of the patients was 84 minutes (60-240 minutes). One patient required postoperative blood transfusion because of a low preoperative hematocrit of about 10 g/dL, but the rest of the patients did not require postoperative blood transfusion. The mean follow-up period was 13 months (6-24 months) and all patients had initial healing of the fracture within 8 weeks after surgery. There were no infections, humeral head necrosis, or any complications related to the internal fixation during the follow-up period. The mean VAS score for postoperative patient pain was 1.5 (0-5). The patients had a mean active forward flexion supination of 147.8° (100°-180°), a mean active lateral external rotation of 45.5° (0-80°), and a mean active internal rotation up to the T10 level (T4-hip) after surgery (Figure 2). In addition, during the follow-up we found that nine patients had mild pressure pain on the surface of the greater tuberosity corresponding to the site of the entry point; in five, the acromion impingement sign could be detected. The JOBE sign, which detects supraspinatus muscle strength, was positive in 6, meaning that the supraspinatus muscle strength was significantly reduced on the affected side compared to the contralateral side. If the supraspinatus, infraspinatus, and teres minor muscles were examined in detail with a force measuring device, it was seen that the average muscle strength of the upper extremity in the 90° forward flexion position in the scapular plane on the healthy side was 14 pounds, while on the affected side it was 9.2 pounds, with a significant difference between the two sides. When the shoulder joint was located on the side of the body, the mean external rotation strength was 14.2 pounds on the healthy side compared to 11.6 pounds on the affected side, and the difference was statistically significant. The mean VAS score for postoperative pain was 1.5 (0-5), the mean ASES score was 81.2 (45.0-98.3), the Constant-Murley score was 85.4 (42.0-100.0), and the UCLA score was Shoulder function was assessed as excellent in 2 patients, good in 16 patients, and poor in 4 patients. On the other hand, all patients expressed satisfaction with the surgical outcome at the follow-up. We performed a multiple linear regression analysis between some preoperative factors that might affect the patients’ final shoulder function and the findings at the postoperative follow-up and the patients’ ASES scores at the final follow-up. The statistical results were as follows: the F-value of the linear regression model was 5.071, p = 0.005, indicating the presence of factors among the enrolled factors that directly influenced the patient’s postoperative ASES score. There was a correlation between the patient’s age (P = 0.001), postoperative supraspinatus muscle strength (P = 0.012), and muscle strength recovery in the infraspinatus and lesser circular muscles (P = 0.0005) and the patient’s final ASES score; in addition the duration of follow-up may also affect the patient’s postoperative shoulder ASES score (P = 0.054). There was no correlation between the time between injury and surgery, whether the injury was on the dominant side, the patient’s gender, the presence of pressure pain on the surface of the greater tuberosity at postoperative follow-up, and the presence of signs of acromion impingement on the injured side and the patient’s final shoulder ASES score (P>0.05). Discussion Significantly displaced proximal humeral fractures have been a more difficult injury to treat, and the reasons for this are generally considered to be the following: 1) 75% of proximal humeral fractures occur in older adults over the age of 60, and more women than men. These patients have extremely osteoporotic bones and are often combined with some systemic diseases, which increases the difficulty of surgical fixation. The common plate used in the past is very difficult to obtain a solid fixation for extremely osteoporotic bones, thus making it difficult to obtain a good function. 2) The local anatomy of the proximal humerus is relatively complex, and if the surgery focuses only on fracture management and ignores the anatomical features of the area, it may cause other problems such as acromion impingement syndrome or rotator cuff injury, thus significantly affecting the function of the shoulder joint. The blood supply is extremely limited and can easily lead to humeral head necrosis due to trauma or medically induced injury. 4) Post-surgical rehabilitation takes a relatively long time, but is essential for the recovery of the patient’s shoulder function. The aim of surgery in such patients is to reconstruct the anatomical form of the proximal humerus and to give good fixation so that the patient can start functional exercises of the shoulder joint early. Given the characteristics of proximal humeral fractures, the application of intramedullary pin fixation in this area has unique advantages. First, PHN can be placed through a smaller incision, avoiding extensive soft tissue dissection and disruption of blood flow to the fracture end, thus improving fracture healing rates; second, for specific types of proximal humeral fractures, PHN fixation can provide sufficient initial stability to allow early initiation of shoulder motion; third, the central splint of intramedullary fixation is more conducive to carrying axial stresses, which is biomechanically more advantageous than other eccentric fixation methods. Fourth, due to its angular stabilization design, PHN can better avoid loosening of the internal fixation after surgery and is more suitable for osteoporotic patients because of its unique proximal rotating blade design, which increases the contact area between the fixation and the bone. Hessmann, M. H. et al. published a biomechanical study comparing the biomechanical strength of the PHN, Philos plate, and the conventional T-plate in fixing two-part surgical neck fractures. The authors collected 24 cadaveric specimens of the proximal humerus, all of which were osteotomized at the surgical neck to create a two-part surgical neck fracture model with an 8 mm long bone defect on the medial side of the surgical neck resulting in an unstable fracture model. These specimens were subjected to axial and torsional loads to test the mechanical properties of the different fixation materials. The authors found that the PHN and Phios plates were stronger than the conventional T-plates in both axial and torsional loading, and that the PHN was the strongest of the three, but not statistically different from the Philos plates. All Philos plates in the disruption test were subjected to increasing axial loads and eventually the plates buckled and failed, whereas failure of PHN fixation occurred in humeral stem fractures distal to the distal locking nail or main nail. From this group of studies we can find that PHN can achieve very reliable fixation for unstable two-part surgical neck fractures. Selection of indications: We currently select patients with 2-part surgical neck fractures as the surgical indication for internal fixation with PHN. In cases where there is a fracture line at the greater tuberosity beyond the entry point, we believe that the soft tissue hinge between the greater tuberosity and the main bone should remain intact as long as there is no significant displacement of the greater tuberosity, and therefore, in such cases, reinforcement of the greater tuberosity with an adjunctive Ethibond wire or titanium cable will provide sufficient initial stability to initiate early functional exercise of the shoulder joint. In cases where the proximal humerus is not cortically intact beyond the intramedullary pin entry point, single-plane fixation with a rotating blade alone provides very limited grip on the humeral head and does not provide adequate stability in the early postoperative period. There is some consensus in the literature on the use of intramedullary pins for the treatment of proximal humeral fractures, and Lin believes that the integrity of the proximal ring is essential to maintain intramedullary fixation, and the authors conclude that this ring is continuous from the humeral head and the lateral greater tuberosity of the cortex. J. Bernard et al. summarized 11 proximal humerus fractures treated with the Polarus intramedullary pin, four of which were two-part surgical neck fractures and the remaining seven were three-part greater tuberosity plus surgical neck fractures. All patients with two-part fractures had no postoperative problems with internal fixation failure, whereas five of the seven patients with three-part fractures had internal fixation failure requiring a second surgery within 6 months after surgery. Therefore, he concluded that the use of intramedullary fixation technique in three- or even four-part fractures has a high failure rate. Intraoperative technical points: 1) Protection of the deltoid muscle: The deltoid is the most important power muscle of the shoulder joint and its function is essential to preserve good shoulder function. During surgery, we generally choose to bluntly separate the deltoid from the anterior angle of the acromion, the gap between the anterior and middle deltoids down to about 4 cm below the anterior angle of the acromion. We routinely suture a fixation line at the lower edge of the separated deltoid gap to prevent damage to the axillary nerve by continuing to split the deltoid downward during the operation. If increased exposure of the proximal humerus is required during the operation, for example, by reinforcing the greater tuberosity with Ethibond wire or titanium cable, the lateral start of the acromion can be severed along the anterior and lateral edges of the anterior and middle deltoid fibers. After the fracture is fixed, we will reconstruct the deltoid very carefully. If the deltoid gap is only separated intraoperatively, the separated gap needs to be closed during reconstruction; if the starting point of the deltoid is cut intraoperatively, the starting point of the deltoid needs to be firmly reconstructed by drilling and crossing the line on the acromion. 2) Entry point of the intramedullary needle: The accurate selection and careful treatment of the entry point of the intramedullary needle is one of the key steps for successful surgery. Most of the relevant literature nowadays believes that the entry point of the intramedullary needle should be located in the depression between the greater tuberosity and the articular surface of the humeral head, and the entry point should be chosen rather slightly inward than outward. If the entry point is too far out, the bone cortex of the greater tuberosity outside the entry point will be less preserved, which will easily split the greater tuberosity during the opening of the medulla, thus affecting the stability of fixation. In addition, an excessively outward entry point may also lead to an inversion deformity of the surgical neck after repositioning, which may affect the patient’s postoperative function. Many humeral surgical neck fractures result in an inversion of the humeral head, causing the greater tuberosity to turn inferior to the acromion, making it difficult to identify a good medially oriented access point. In these cases, it is often necessary to extremely pronate and posteriorly extend the shoulder joint so that the correct pulp opening point on the lateral aspect of the humeral head is as far outward and forward as possible to avoid the acromion. If this is still not sufficient to allow us to successfully identify the pulp opening point due to the inversion of the humeral head, it is necessary to temporarily pry the humeral head from the folded end with a periosteal striker or a single hook to correct the inversion deformity and reveal the correct entry point to facilitate pulp opening (Figure 4). One problem that may arise from a paralleling intramedullary fixation is the placement of the internal fixation through the supraspinatus tendon, which may lead to painful symptoms in the shoulder joint due to rotator cuff injury or rotator cuff tendonitis and subacromial impingement secondary to the placement of the internal fixation. We believe that the caudal end of the intramedullary pin should be placed at a depth at least flush with the surrounding articular surface or even buried below the articular surface of the humeral head to avoid impingement. Since the PHN intramedullary pin is made of titanium, we usually inform the patient preoperatively that the fixation will remain in the body for a long time and that it is not recommended to be removed, so there is no need to consider that the intramedullary pin is placed too deep to prevent its removal. We also emphasize splitting the tendon fibers in the direction of the supraspinatus muscle before fixation and carefully repairing the incised supraspinatus tendon after fixation. However, even after such careful treatment, we found that nine patients had mild pressure pain on the surface of the greater tuberosity near the intramedullary needle entry point during follow-up, and five of them had different degrees of acromioclavicular impingement. six patients had positive JOBE signs, suggesting weakness of the supraspinatus muscle on the affected side. The mean values of supraspinatus and posterior infraspinatus and teres minor muscle strength were worse than those of the healthy side in all patients and the difference was significant. We believe that the reasons for these results may be multiple. First, the follow-up time of this group of patients was not long enough, with a mean of 57 weeks and a minimum of 25 weeks, and it is possible that with the prolongation of postoperative rehabilitation, many patients may have further recovery of rotator cuff strength. In addition, 15 of the cases in this group were involved on the non-dominant side, which was considerably more than the cases with involvement on the dominant side. Therefore, in many patients, the strength of the rotator cuff on the non-dominant side was compared with that on the dominant side, which would normally be slightly less than the latter. However, it is undeniable that the weakness of the painful shoulder joint caused by the paralleling humeral intramedullary pin is an inherent weakness of this internal fixation design, and many reports in the literature have noted this problem.John Crates reported on 73 patients treated with paralleling intramedullary pins for humeral stem fractures, and a mean follow-up of 12 months after surgery showed a total of 7 patients with shoulder joint pain.