1. Inclusion and exclusion criteria.

This is a retrospective single-center study. Approved by the Institutional Review Board of Xi'an Honghui Hospital, we retrospectively screened FNF patients treated by FNS at our hospital from January 2019 to December 2021. Inclusion criteria: (1)patients treated by FNS and at least 6 months after surgery for FNF,(2) closed femoral neck fracture,(3) age less than 70 years old,(4) complete radiographic data; Exclusion criteria: (1)proximal femur fracture of the ipsilateral limb,(2) pathological fracture,(3) patients with immature bones, (4)patients with FHN, (5)patients who drink a lot of alcohol over the long term, (6) patients using hormonal drugs, (7)patients with significant internal diseases including cardiovascular, cerebrovascular, pulmonary system, liver system, renal system, etc.

After screening, 109 patients with 109 FNF are included in this study. Among them, 59 are male and 50 are female. The mean age at initial treatment was 49.71 ± 9.84 years, ranging from 27 to 67 years, with a follow-up time range of 6 to 29 months.

2. Date and date collection

First, we collect the general information, length of surgery, volume of blood loss, and length of incision about the patient from the electronic medical record system.

Second, we download the preoperative and postoperative X-rays from the electronic imaging system, which are used to obtain the Garden classification, Pauwels classification, Tip-apex distance (TAD), Parker ratio, Pauwels angle, and the difference in neck stem angle between the healthy and the injured side after surgery, as well as the length of femoral neck shortening.

Third, we plot the downloaded X-rays, then measure the relevant data and record. The data collection was accomplished by the same author.

The Garden classification is used to evaluate whether the fracture is stable [5] and the Pauwels classification, Pauwels angle, are used to evaluate the shear stress of the fracture [6]. The TAD and the Parker ratio are used to evaluate the position of the inserted implant, the angle of neck stem and the length of femoral neck can be used to evaluate the quality of the reduction.

3. Measurement methods of data

The FNS we used came from Tianjin Zhengtian Medical Equipment Co. Product Model:CL010000. It consists of bolt, anti-rotation screw, bone plate, and locking head screw. The diameter of the bolt is 10 mm, which is used to correct enlarged errors in the measurement [7] and establish a standardized measurement protocol. The download X-rays of the patient are edited on the software WPS Office.

Measurement of positive X-Rays: First, draw a circle matching the diameter of the femoral head [8], make two tangents to the circle, and then make the vertical lines of these two tangents across the tangent point, the intersection of the two vertical lines being the center of the circle Point O. (Because most closed fracture restorations are not anatomical, the horizontal line of the patient's greater trochanter does not cross the center of the circle, so it is more appropriate to use this method to find the center of the circle.)

Second, draw a linear segment perpendicular to both edges of the bolt, which is the diameter of the bolt, and make the vertical bisector of the linear segment, which is the axis of the bolt.

Third, do the vertical line of the bolt axis over the center of the circle Point O, intersecting the femoral head at two points, the inner point is point A, the outer point is point C, and the intersection with the bolt axis is point B. AB/AC × 100 is the Parker ratio of the positional X-rays.

Fourth, connect the two intersections of the circle and the femoral neck, and make the vertical line of this line through the center of the circle Point O, which is the axis of the femoral neck. The intersection of the axis of the femoral neck and the femoral head is point E, the intersection of the axis of the bolt and bolt’s top is point F. The length of EF is positive TAD.

Fifth, the four-point method is used to make the axis of the femoral stem. The axis of the femoral stem interacts with the axis of the femoral neck at point D, OD is the length of the femoral neck, and ∠D is the angle of neck stem, as shown in Fig. 1(a). The length of the femoral neck on the healthy side is drawn in the same way, as well as the angle of neck stem, as shown in Fig. 1(b).

Sixth, make a vertical line of the femoral neck axis across the upper edge of the femoral head and intersect the femoral neck axis at point G, then make a parallel line of fracture line through the point G. ∠G is the Pauwels angle, as shown in Fig. 1(d).

Measurement of lateral X-ray: The lateral X-rays are drawn in the same way as the positive X-rays, the point where the bolt axis intersects its vertical line is point B, the intersection of the vertical line with the anterior side of the femoral head is point C, and the intersection of the posterior side is point A, and AB/AC×100 is the Parker ratio of the lateral position. The distance between point E and point F is the TAD of the lateral radiograph, as shown in Fig. 1(c). After drawing, the length and angle are measured by the software ruler.

TAD = measurement of positive TAD× (bolt real diameter / positive bolt measurement diameter) + measurement of lateral TAD× (bolt real diameter / lateral bolt measurement diameter)

Parker ratio = AB/AC×100

Length of femoral neck shortening = difference in measured of the length of femoral neck between the injured and healthy side × (true diameter of the bolt / measured diameter of the positive bolt)

The difference of neck stem angle = the neck stem angle of health side- the neck stem angle of injured side

Radiologically, it is believed that there is a 27% chance of HR at TAD > 30 mm, a 2% chance of HR at 25 mm ≤ TAD ≤ 30 mm, and a 0% chance of HR at TAD < 25 mm [9]. Meanwhile, Parker ratio ≥ 66 is considered that the position of the bolt is implanted upper in the positive position and posterior in the lateral position; for Parker ratio ≤ 33, the position of the bolt is implanted lower in the positive position and anterior in the lateral position [10].

There are three grades about the shortening of femoral neck, slight shortening (≤ 5 mm), medium shortening (6–10 mm), and severe shortening (> 10 mm) [11]. Whether there is an association between the length of the femoral neck shortening and changes in the angle of neck stem and complications has not been mentioned in the previous literature.

4. Statistical analysis

Data are analyzed by IBM SPSS Statistics version 25.0, and the level of statistical significance is set at P < 0.05. The data are divided into two groups, normal and complications, and correlated factors leading to complications are examined by using binary logistic regression - forward LR method, if factors are not correlated which shown by binary logistic regression, then using Fisher exact test, Pearson test, to observe the difference between the groups.