Effect of Blood Transfusion on Short- and Long-Term Outcomes in Oral Squamous Cell Carcinoma Patients Undergoing Free Flap Reconstruction

Purpose: To analyze the short- and long-term effect of perioperative blood transfusion (PBT) in patients undergoing surgical treatment for oral squamous cell carcinoma (SCC). Methods: Patients undergoing free flap reconstruction were retrospectively enrolled and divided into two groups based on the implementation of PBT. Flap revision, surgical site infection (SSI), flap failure, overall survival (OS), and disease-specific survival (DSS) were compared between the two groups. Results: In 170 patients with PBT, 10 (5.9%) flaps required exploration revision, SSI occurred in 18 (10.6%) patients, and flap necrosis was noted in 6 (3.5%) patients. These rates were comparable to those in patients without PBT. The two groups had similar DSS rates, but the 5-year OS rates were 49 and 59% in patients with PBT and without PBT, respectively. This difference was significant. Patients with 4 units of PBT had OS rates comparable to those of patients with >4 units of PBT. A Cox model confirmed the fact that the decrease in OS was independent of PBT. Conclusion: In patients with free flap reconstruction for oral SCC, PBT did not increase the short-term complication rate or cancer-linked mortality. However, it was related to an elevated overall risk of death.


INTRODUCTION
Oral squamous cell carcinoma (SCC) is the most common malignancy in cancers of the head and neck (1), usually presenting as advanced-stage disease on initial treatment. Due to its site specificity, advanced-stage patients sometimes complain of poor appetite, pain, and dysphagia (2), and a number of patients have anemia with different degrees of severity. Additionally, in these patients, a free flap reconstruction is frequently required, and intraoperative blood loss is significant (3). Therefore, perioperative blood transfusion (PBT) is not uncommon in patients receiving free flap reconstruction for oral SCC.
Immunosuppression induced by PBT was first discovered by Opelz et al. (4) in a renal transplant. A series of studies thereafter have proven the adverse influence of PBT on oncologic survival in cholangiocarcinoma (5) and liver (6), bladder (7), pancreatic (8), gastric (9), and lung cancer (10). However, the effect on oral SCC remains controversial. A summary by Szakmany et al. (11) reported that some studies reported no significant effect of PBT, while some noted a decreased survival rate related to PBT in a dose-dependent manner. Therefore, in the current study, we aimed to analyze the short-and long-term effect of PBT in patients undergoing free flap reconstruction for oral SCC.

Ethical Consideration
The institutional research committee of Zhengzhou University approved our study, and all participants provided written informed consent for medical research prior to the initial treatment. All experiments were performed in accordance with the relevant guidelines and regulations.
From January 2010 to December 2018, all patients undergoing free flap reconstruction for untreated primary oral SCC were retrospectively enrolled. Thereafter, data regarding demography, systemic disease, including hypertension, diabetes and cardiovascular and cerebrovascular diseases, TNM stage based on the 8th AJCC classification, PBT, ASA status, Eastern Cooperative Oncology Group (ECOG) status, preoperative serum albumin level, postoperative pathology, treatment, and follow-up were acquired and analyzed.
The perioperative period was defined as the timespan between the start of the patient's hospitalization and the time they left the surgery department. All patients underwent systemic preoperative examinations, including ultrasound, CT, MRI and/or PET-CT imaging. Pathological sections were reviewed by at least two head and neck pathologists. Perineural invasion (PNI) was considered present if tumor cells were identified within the perineural space and/or nerve bundle. Lymphovascular infiltration (LVI) was considered positive if tumor cells were noted within the lymphovascular channels (12)(13)(14)(15)(16). The depth of invasion was measured from the level of the adjacent normal mucosa to the deepest point of tumor infiltration, regardless of the presence or absence of ulceration (12). The indications for adjuvant treatment included neck lymph node metastasis, PNI, LVI, high tumor stage, positive margins, and extracapsular spread (12). Following discharge, patients were followed every 3 months during the first year, every 6 months during the second year, and once a year after the second year (17).
In general, our blood units used were all leukocyte-depleted, and a preoperative hemoglobin level of <60 g/L required to increase it to at least 80 g/L. PBT was selectively performed under the consideration of the surgeon and the anesthesiologist if the hemoglobin level was 60-90 g/L. The choice to undertake an intraoperative blood transfusion depended on the preoperative hemoglobin level, intraoperative blood loss, and the patient's general condition. Postoperative blood transfusion was performed mainly based on the latest hemoglobin level.
All patients were managed according to the same method. The postoperative flap was observed hourly during the first 24 h and then every 4 h for the next 3 days. If there was a suspicion of vascular events, an exploration operation was immediately performed. All patients received prophylactic antiinfective therapy with cephalosporin 30 min prior to incision, which was continued for at least 3-5 days following surgery. Nasogastric nutrition was routinely administered until patients could eat by mouth.
Short-term outcomes included flap revision, surgical site infection (SSI), and flap necrosis. Flap revision was defined as a surgical intervention complicated by a postoperative vascular event. SSI was defined as wound dehiscence with salivary leak or purulent drainage (18), and flap necrosis was defined as total flap failure. The association between short-term outcomes and PBT was analyzed using a Chi-square test.
Long-term outcomes included overall survival (OS) and disease-specific survival (DSS). The survival time of OS was calculated from the date of operation to the date of death from any cause or the last follow-up. Similarly, the survival time of DSS was calculated from the date of operation to the date of cancercaused death or the last follow-up. The association between longterm outcomes and PBT was analyzed using a Kaplan-Meier method and Cox model.
All statistical analyses were performed with SPSS 20.0, and a value of p < 0.05 was considered to be significant. Positive margin occurred in 63 (14.9%) patients. All patients underwent a neck dissection, and pathologic neck metastasis was confirmed in 221 (52.2%) patients. Patients with PBT had a poorer ECOG status (p = 0.040), more systemic disease (p = 0.004), and a poorer ASA status (p = 0.003) than patients without PBT. The two groups had no significant difference regarding other clinical and pathologic data (all p > 0.05) ( Table 1).
During the follow-up period lasting a mean of 4.9 years, in patients with PBT, 157 patients received postoperative of whom patients died from the cancer. The 5-year OS and DSS rates were 59 and 60%, respectively. The OS difference between the two groups was significant (Figure 1, p = 0.041), but the DSS difference between the two groups was not significant (Figure 2 and Table 3, p = 0.799). A further Cox model analysis confirmed the significance of PBT in predicting the OS ( Table 4).
The association between the amount of PBT and OS was further analyzed in the 170 patients. In the 100 patients receiving 4 units of PBT, the 5-year OS rate was 50%; in the 70 patients receiving more than 4 units of PBT, the 5-year OS rate was 49%; this difference was not significant (Figure 3, p = 0.839).

DISCUSSION
The most significant finding in the current study was that PBT did not increase the risk of flap revision, SSI, flap necrosis, or cancer-caused death. However, it was associated with decreased OS in a dose-dependent manner. This provided powerful clinical value: although PBT was sometimes inevitable, a restrictive transfusion policy must be ensured.
Postoperative complications were an important aspect in perioperative management. Innutrition, including anemia, had significant negative influence on wound healing. One main goal of PBT was to improve low hemoglobin levels. It was interesting, as such, to analyze the change in the true incidence of postoperative complications after PBT. Böck et al. (19) (18) assessed the outcome of PBT in 167 patients with head and neck cancer undergoing free tissue transfer. The rates of wound infection in patients who received 0, 1, 2, or 3 units blood were 13.3, 21.2, 33.3, and 31.2%, respectively. In addition, patients receiving ≥2 units had higher wound infection rates than patients receiving <2 units, but there was no difference regarding flap revision or failure between the two groups. No other studies on the association between PBT and postoperative complication in head and neck cancer were available. The current study had the largest sample size to the best of our knowledge, and our findings partially supported the conclusions of Danan et al. (18). However, we noted the PBT did not increase the SSI rate. This difference could be explained by the different distribution of ASA class: about half of the patients in the research performed by Danan et al. (18) were staged 3/4. Oncologic survival was another important aspect of the assessment of the effect of PBT. PBT was more likely to occur in patients with high ASA score and low preoperative hemoglobin levels (22), and PBT-related immunosuppression was a more involved subject of discussion. It was characterized by increased suppressor T-cell activity and appeared to be dependent on the number of transfusions. As such, theoretically, PBT could both affect the OS and DSS. In a study conducted by Szakmany et al. on 559 patients with oral or oropharyngeal SCC (11), after adjusting for relevant prognostic factors in a Cox regression, the hazard ratio for patients receiving ≥3 units of PBT relative to those not transfused was 1.52 for cancer-caused death and 1.52 for overall mortality. However, there was no DSS difference between patients without PBT and with 1-2 units of PBT. Fenner et al.   death after ≥3 units transfused was 5.79. Therefore, there clearly remains a certain degree of controversy with regard to the association between PBT and prognosis in the current literature. There were at least three aspects contributing to the variation. First, different objects were enrolled, and PBT was frequently required in advanced-stage patients, patients undergoing free flap reconstruction, or patients with severe anemia. Furthermore, PBT-related immunosuppression was usually induced by cells, biosoluble factors, and microchimerism, but the transfused allogeneic white blood cells were responsible for the majority of observed immunosuppression effects (26). In addition, the blood products varied across studies. Third, it was important to differentiate whether PBT was required for preoperative anemia, intraoperative blood loss, or both. In the current study, it was noted that PBT was related to decreased OS but not DSS in a dose-dependent manner. This finding was interesting, and partially explained by the fact that more patients receiving PBT had a poorer ECOG status, meaning they were more likely to die of other non-cancer-related causes. We only included patients undergoing free flap surgeries, and the blood products were leukocyte-depleted. This said, comparing the outcomes among patients who received PBT for different reasons remained difficult, especially in the context of grouping patients with moderate anemia requiring PBT. More high-quality studies were needed to clarify these questions.
Limitations of the study must be acknowledged. First, a selective bias due to the retrospective nature of the study was inevitable. Second, there were other short-term indicators that were not analyzed in current study, which could potentially result in interesting findings.
In summary, in patients with free flap reconstruction for oral SCC, PBT did not increase the short-term complication rate or cancer-linked mortality; however, it was related to elevated overall death risk.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary Material, further inquiries can be directed to the corresponding authors.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Zhengzhou University institutional research committee, and all participants provided written informed consent for medical research prior to the initial treatment. All experiments were performed in accordance with the relevant guidelines and regulations. The patients/participants provided their written informed consent to participate in this study.