The results of this study showed that the VBL and HBL of OP patients was significantly higher than non-osteoporosis patients. This demonstrates that OP significantly increases perioperative blood loss in patients with lumbar degenerative diseases when undergoing Mis-TLIF. It might be that lumbar degenerative diseases patients with OP have reduced bone mass and trabecular thinning, and fracturing. When the screws are passed through the cancellous bone via the pedicle channel, the trabecular structure is destroyed [11], leading to blood in the vertebrae permeating along the pedicle channel and screw thread, which significantly increases blood loss. Internal fixation of the spine requires healthy bone storage, and patients with OP have high bone loss, which is likely to cause complications such as pedicle fracture, vertebral compression fracture, internal fixation failure and adjacent vertebral fracture, and even cause greater blood loss [12, 13].

HBL occurs during the perioperative period in most orthopedic surgery. The average HBL in total hip arthroplasty was 471 ml, accounting for 26% of the TBL, according to Sehat et al. [14]. In spinal surgery, Smorgick et al. [15] showed that the average total loss of blood in posterior spinal fusion surgery was 1439 ml, of which HBL accounted for 42%. Xu et al. [16] showed HBL of 362.8 ml in a patient undergoing TLIF for degenerative lumbar disease, accounting for 47% of the TBL. Similarly, in our study, for lumbar degenerative diseases patients undergoing Mis-TLIF surgery, the HBL was 173.04 ml, accounting for 31.4% of the TBL. However, for patients with OP, HBL accounted for as much as 46.1% of the TBL, which was larger than expected. Mis-TLIF removed some of the paravertebral muscles, ligaments and some of the lamina, leading to excessive perioperative VBL. Under this circumstance, HBL is often ignored in the assessment of perioperative bleeding, especially in cases that are complicated with OP and other conditions. Excessive HBL affects patients’ general condition, resulting in postoperative hemodynamic instability, blood transfusion, recovery delay and other problems [8]. Our study also showed that the volume of allogeneic blood transfusions of OP patients was significantly higher than in patients with non-osteoporosis.

We included preoperative regular anti-osteoporosis treatment in the analysis. Our results showed that regular anti-osteoporosis treatment for at least 6 months before surgery was an independent risk factor for HBL in OP patients and negatively correlated with the amount of HBL. It might be that regular anti-osteoporosis increases bone mineral density (BMD) and reduces the risk of fractures in postmenopausal women with OP [17]. Even some authors believe that patients with OP, multiple-segment fixation, or age > 65 years, with a dual energy X-ray absorptiometry outcome of T ≤ −1.0, whose surgery can be postponed, are recommended to receive at least 6 months of anti-osteoporosis treatment before surgery. In other patients whose operations cannot be postponed, immediate anti-osteoporosis treatment is needed after surgery [18]. Therefore, it is necessary that spinal surgeons should lay more emphasis on OP during the perioperative period.

We also found that the use of TXA during or after spinal surgery was an independent risk factor in our study, and that the total amount of HBL was negatively correlated with TXA. According to Willner et al. [19], patients who received tranexamic acid had a 49% reduction in blood loss and required 80% fewer blood transfusions compared with patients treated with placebo. Large doses of tranexamic acid provide an effective and inexpensive method for reducing blood loss during spinal surgery. Ren et al. showed that topical tranexamic acid can reduce HBL without increasing the risk of deep vein thrombosis, pulmonary embolism, and other complications [20]. On the basis of the above examples, we think the application of tranexamic acid can effectively reduce HBL.

It is controversial that age, BMI, subcutaneous fat thickness and muscle thickness are related to HBL. Madsen et al. [21] have concluded that advanced age is an independent risk factor for increased HBL. In contrast, Wu et al. [22] found that age was not related to HBL. Zhou et al. [5] reported that muscle thickness was the key factor in predicting HBL in Mis-TLIF, whereas subcutaneous fat thickness and BMI were not risk factors for HBL. In our study, age, BMI, subcutaneous fat thickness and muscle thickness showed significant differences between the Osteoporosis and Non-osteoporosis groups, but were not independent risk factors for HBL in OP patients. This may be because patients with lumbar degenerative disease combined with OP are older and have thinner muscle and fat.

Our study showed that Hctpost was less in the Osteoporosis group, while Hctpre was no different between the two groups, which may be due to postoperative fluid dilution and blood loss leading to more change in Hct [23]. On the one hand, all patients were free from anemia preoperatively, so the Hctpre of the two groups was no different. On the other hand, OP patients had more blood loss than non-osteoporosis patients had, which resulted in a significant reduction in Hctpost. Some studies have shown that postoperative fluid dilution could contribute to Hct change [24], which might explain why Hctpost was an independent risk factor for HBL in OP patients. In our study, T-score was also an independent risk factor. It is likely that lower T-scores indicate more severe osteoporosis and, consequently, greater HBL.

There were some limitations to this study. First, this study was retrospective and had a small sample, which failed to reflect most people’s real circumstances. Data biases existed, so a larger sample size is needed for validation. Second, some patients had a large amount of blood leakage from the postoperative wound, which was not counted as postoperative blood loss, and the reliability of the conclusion was affected when calculating blood loss using hematocrit after surgery.