Dynamic Analysis of Perioperative Hidden Blood Loss in Intertrochanteric Fractures

Abstract

To analyze the dynamic variation in perioperative hidden blood loss in patients with intertrochanteric fracture. From January to December 2017, 79 patients with intertrochanteric fracture were treated with proximal femoral nail antirotation. Serial complete blood count assays were performed consecutively in the 3 days after admission, on the day of surgery, and 7 days postoperatively. Blood loss during surgery, postoperative drainage, and perioperative blood transfusion volumes were recorded. Dynamic changes in hemoglobin (Hb) prior to surgery were recorded and compared between males and females. Patients were divided into the no blood transfusion group, the 400-mL blood transfusion group, and the 800-mL blood transfusion group depending on the volume of perioperative blood transfusion. Total and hidden blood loss were separately calculated according to the Gross equation. Lowest mean Hb values occurred on day 2 after admission among men (104.8 g/L) and on day 3 after admission among women (98.6 g/L). The average Hb decrease was 11.4 g/L, 11.8 g/L, and 8.9 g/L in the no, 400-mL, and 800-mL blood transfusion groups, respectively. The lowest Hb value occurred on postoperative day 2. Hemoglobin increased on postoperative day 3 and stabilized by day 6. In the no blood transfusion group, the average total blood loss was 406.0 ± 255.6 mL, 628.3 ± 267.2 mL, and 759.7 ± 322.1 mL in the no blood transfusion, 400-mL blood transfusion, and 800-mL blood transfusion groups, respectively, and hidden blood loss was 326.0 ± 246.6 mL, 512.1 ± 247.3 mL, and 596.1 ± 306.9 mL, respectively. Perioperative hidden blood loss occurred prior to surgery for intertrochanteric fracture and ended on postoperative day 2.

Introduction

Although surgical and anesthetic techniques have made great progress, the mortality rate of patients with intertrochanteric fracture has not significantly changed in the last 40 years.¹ An important risk factor for perioperative death is anemia. Many patients have varying degrees of anemia in the perioperative period. Identifying and rectifying perioperative anemia earlier could facilitate appropriate medical optimization and reduce mortality.² The aim of our study was to improve understanding of this issue by monitoring dynamic variations in perioperative hidden blood loss and hemoglobin (Hb) levels in patients undergoing surgery for intertrochanteric fracture.

Patients and Methods

From January to December 2017, a total of 109 surgeries for proximal femoral nail antirotation were performed in our hospital. Inclusion criteria were as follows: (1) intertrochanteric fracture patients >65 years of age and admission on the day of fracture occurrence and (2) platelet count, prothrombin time, partial thromboplastin time, and international normalized ratio within normal ranges. Exclusion criteria were as follows: (1) pathological fracture; (2) serious cardiac or respiratory disease requiring conservative treatment; (3) congenital or acquired coagulopathy; (4) history of thromboembolic disease such as cerebral infarction, pulmonary embolism, myocardial infarction, deep vein thrombosis, or recent thrombophilia with chronic anticoagulant drugs; and (5) hemorrhagic disease developed during hospitalization.

A total of 79 patients were included in our study. We eliminated several of the original 109 cases because they could have affected the authenticity of the results, as follows: 7 patients who were not admitted to hospital on the day of fracture occurrence, 12 who completed surgery within 48 hours, 3 cases of congenital or acquired coagulopathy, 6 with chronic anticoagulant drugs because of thrombus history, and 2 who developed hemorrhagic diseases during hospitalization (1 cerebral hemorrhage and 1 gastrointestinal hemorrhage). All patients received low-molecular-weight heparin (subcutaneous enoxaparin, 40 mg, once daily) from the time of admission (pre- and postoperatively). The day of admission was recorded as day 0. Patients had daily consecutive complete blood count assays for 3 days after admission, on the day of the operation, and on the first 7 days postoperatively. The volumes of visible blood loss in the operating room, postoperative drainage, and perioperative blood transfusion were recorded. The 79 patients were divided into male and female groups; the dynamic changes in Hb in each group prior to surgery were recorded and compared.

All of our patients followed the new blood transfusion guidelines, which stated that transfusions should occur when Hb concentrations were less than 8 g/dL.³ Patients were further divided into the no blood transfusion group, the 400-mL blood transfusion group, and the 800-mL blood transfusion group depending on the volume of perioperative blood transfusion. Values of perioperative variations in Hb and hematocrit (Hct) were recorded from day 1 after admission to the lowest day postoperatively in each group. Hidden blood loss and total blood loss were calculated according to the Gross formula. The dynamic changes in Hb level from the day of surgery to postoperative day 7 were analyzed for the no blood transfusion group.

Approval for our study was provided by the ethics board of the Third Hospital of Hebei Medical University [clinical trial registration code: 2016-016-1], and all patients provide their written informed consent for participation.

Calculation of Hidden Blood Loss

The patient’s blood volume (PBV) was calculated as follows⁴:

$${\text{PBV = k}}_{\text{1}}\text{× height }{\left(\text{m}\right)}^{\text{3}}{\text{+ k}}_{\text{2}}\text{× weight }\left(\text{kg}\right){\text{ + k}}_{\text{3}},$$ 1

where k₁ = .3669, k₂ = .03219, and k₃ = .6041 for men and k₁ = .3561, k₂ = .03308, and k₃ = .1833 for women.

Multiplying the PBV by the Hct yields the total red cell volume. Any change in red cell volume can therefore be calculated from the change in Hct. Total red blood cell (RBC) volume loss = PBV × (Hct_pre-op − Hct_post-op).⁵

The theoretical total blood loss was determined according to total RBC loss and Hct_pre-op. The formulae used were as follows:

$${\text{Theoretical total blood loss = total RBC loss/Hct}}_{\text{pre-op}}\text{.}$$

$$\text{Perioperative actual blood loss =Theoretical total blood loss}+\text{transfusion}=\text{the hidden blood loss}+\text{visible blood loss}$$

$$\text{Hidden blood loss=theoretical total blood loss}+\text{transfusion - visible blood loss}$$

When transfusion was performed, a unit of red cell concentrate containing the standard 200 mL of red blood corpuscles was used.

Statistical Methods

Statistical analysis was performed using SPSS 16.0 statistical software (SPSS Inc., Chicago, Illinois). Means are presented as mean (standard deviation). Numerical data were compared by t test. The significance level used for all tests was P < .05.

Results

This study comprised 37 males and 42 females with a mean age of 74.5 years (range, 67-82 years), average height of 1.61 m (range, 1.50-1.73 m), and a mean body weight of 64.4 kg (range, 50.3-81.1 kg). Fractures classified according to the orthopedic trauma association (AO/OTA) classification included 31 A1 fractures in 19 patients, 31 A2 fractures in 38 patients, and 31 A3 fractures in 22 patients. A total of 31 patients did not receive a blood transfusion, 26 received a transfusion of 400 mL, and 22 received a transfusion of 800 mL during the perioperative period. All cases of severe anemia were corrected promptly during the study, and no patient died of anemia. Patient demographic characteristics are shown in Table 1.

Table 1.

Patient Demographic Characteristics.^a

Variables	Value
Patients	79
Male	37
Female	42
Age, years (SD)	74.5 (6.3)
Height, m (SD)	1.61 (0.12)
Weight, kg (SD)	64.4 (11.8)
BMI, kg/m2 (SD)	24.8 (4.52)
OTA classification (%)
31A1 fractures	19 (24.05%)
31A2 fractures	38 (48.10%)
31A3 fractures	22 (27.85%)
Blood transfusion, mL (%)
0	31 (39.24%)
400	26 (32.91%)
800	22 (27.85%)

Abbreviations: BMI, body mass index; SD, standard deviation.

^a n = 79.

Mean Hb values on day 1 after admission were 111.4 g/L for men and 106.8 g/L for women, far below normal values. The lowest Hb values occurred on day 2 after admission in the male group and on day 3 after admission in the female group (104.8 g/L and 98.6 g/L, respectively). The mean serial Hb results and dynamic variation trends for the male and female groups on the first 3 days after admission are illustrated in Figure 1.

Figure 1.

Dynamic variation in mean hemoglobin prior to surgery in patients with intertrochanteric fractures.

Comparisons of blood loss in each group are shown in Table 2. In the no blood transfusion group, the total blood loss was 406.0 ± 255.6 mL including overt blood loss of 80.0 ± 43.1 mL and hidden blood loss of 326.0 ± 246.6 mL; hidden blood loss was 85.1% of the total blood loss. In the 400-mL blood transfusion group, the total blood loss was 628.3 ± 267.2 mL including overt blood loss of 116.2 ± 57.6 mL and hidden blood loss of 512.1 ± 247.3 mL; hidden blood loss accounted for 78.9% of the total blood loss. In the 800-mL blood transfusion group, the total blood loss was 759.7 ± 322.1 mL including overt blood loss of 163.6 ± 62.4 mL and hidden blood loss of 596.1 ± 306.9 mL; hidden blood loss accounted for 73.9% of the total blood loss.

Table 2.

Comparison of Hb/Hct Levels and Blood Loss in the No Transfusion, 400-mL Transfusion, and 800-mL Transfusion Groups.

Parameter	0 mL, n = 31	400 mL, n = 26	800 mL, n = 22	P Value
Hb, g/L
Pre	112.1 (8.9)	109.0 (10.3)	104.5 (6.8)	.106
Post	100.7 (5.6)	97.2 (9.5)	95.5 (5.0)	.022
-	11.4 (8.6)	11.8 (8.9)	9.0 (8.5)	.004
Hct, %
Pre	33.5 (2.7)	32.5 (3.1)	31.2 (2.0)	.492
Post	30.1 (1.7)	29.0 (2.8)	28.5 (1.5)	.038
-	3.4 (1.7)	3.5 (2.7)	2.7 (2.5)	.015
Overt blood loss, mL	80.0 (43.1)	116.2 (57.6)	163.6 (62.4)	.009
Hidden blood loss, mL	326.0 (246.6)	512.1 (247.3)	596.1 (306.9)	.011
Total blood loss, mL	406.0 (255.6)	628.3 (267.2)	759.7 (322.1)	.005
Hidden/total, %	85.1	78.9	73.9	.023

Abbreviations: Hb, hemoglobin; Hct, hematocrit; Pre, postoperative; Post, postoperative; –, post-pre; Hidden/Total, hidden blood loss/total blood loss.

The mean Hb on the day of operation in the no blood transfusion group was 106.8 g/L; the lowest Hb value (100.7 g/L) appeared on postoperative day 2. Hemoglobin increased on postoperative day 3 and stabilized by day 6. The mean decrease in Hb between samples taken on the day of surgery and those taken on postoperative day 2 was 6.1 g/L for patients with intertrochanteric hip fracture. The mean serial Hb results and dynamic variation trend for the no blood transfusion group between the day of surgery and postoperative day 7 are illustrated in Figure 2.

Figure 2.

Postoperative trend in mean hemoglobin dynamic variation.

Discussion

Patients with hip fractures have a high perioperative mortality.⁶ For example, European patients with hip fracture have a 30-day mortality in excess of 10%⁷ and a 1-year mortality of more than 25%.⁸ Perioperative morbidity in surgical procedures is often multifactorial, and anemia is one such important factor. This increases the incidence of pulmonary edema and cerebral edema and breaks the balance of oxygen supply and demand in the tissues, causing other problems.⁹ In a case–control study of 125 surgical patients who refused blood transfusions, operative mortality was inversely related to the preoperative Hb level, increasing from 7.1% for patients with levels of greater than 10 g/dL to 61.5% for those with levels of less than 6 g/dL.²

Previous work has put particular emphasis on analyzing the hidden blood loss related to surgery¹⁰ but has not addressed the hidden blood loss that occurs preoperatively. The serial Hb results from our study show that the mean decrease in Hb between day 1 after admission and the day of surgery was 6.6 g/L for men and 8.2 g/L for women. Therefore, we conclude that hidden blood loss occurred prior to surgery. Smith et al¹¹ demonstrated that the hidden blood loss associated with initial trauma was up to 400 mL by day 3 after fracture. We also found that the lowest Hb value occurred on day 2 for men and day 3 for women, indicating that the Hb value on the day of admission does not accurately reflect BV status at the time of surgery. We suggest rechecking routine blood counts several times after admission, especially in older female patients.

Previous studies have measured total blood loss as intraoperative hemorrhage plus postoperative drainage, in other words, the visible blood loss. We observed a mean decrease in Hb between admission and the lowest Hb prior to surgery of 11.4 g/L in the no blood transfusion group, 11.8 g/L in the 400-mL blood transfusion group, and 9.0 g/L in the 800-mL blood transfusion group, as illustrated in Table 2, which was greater than the visible blood loss. This “invisible” blood loss is the hidden blood loss and as a proportion of total blood loss it is substantial. Hidden blood loss as a percentage of total blood loss in our study was 85.1%, 78.9%, and 73.9% in the 3 groups, respectively, and was up to 3.5 to 4.4 times more than that observed during the surgical procedure. Hidden blood loss has a direct effect on the perioperative management of patients with intertrochanteric fracture. A better understanding of hidden blood loss is important to facilitate appropriate medical optimization.

Hidden blood loss can result from extravasation of blood from the surgical wound and fracture into the tissues. Previous experiments using labeled RBCs demonstrated that this hidden blood loss results from perioperative bleeding into tissue compartments.¹² Extravasated blood no longer circulates throughout the body, resulting in a reduced effective circulating BV and decreased Hb. Other studies have shown that hidden blood loss can be partly attributable to hemolysis caused by traumatic stress, surgery, and anesthesia.¹³ For example, Foss and Kehlet¹⁰ demonstrated that perioperative hidden blood loss was associated with the type of surgery, treatment with aspirin, intraoperative hypotension, and gastrointestinal bleeding or ulceration. Frequent perioperative measurements of Hb also exacerbate RBC loss.¹⁴

Patients often have severe anemia after surgery, and this is not always confined to postoperative day 1. We therefore conjecture that the Hb result on postoperative day 1 is not necessarily representative of the true lowest value for that patient. Because blood transfusion greatly influences the natural postoperative trends, we included 31 patients who did not receive blood transfusion postoperatively and made statistical descriptions of the trends in Hb changes (Figure 2). Postoperative Hb values decreased, with the lowest value occurring on postoperative day 2. Compared to the day of surgery, the mean Hb decrease was 6.1 g/L. After that, Hb increased beginning on postoperative day 3 and stabilized by day 6. This pattern highlights the need for orthopedic staff to be vigilant concerning the risk of anemia on postoperative day 2. It is essential to measure Hb on this day, and we also suggest that routine blood counts should be rechecked many times after surgery, especially in patients with blood transfusion. For older people with poor mobility and decreased physiological compensation, finding and treating perioperative anemia promptly can significantly improve survival.¹⁰

Conclusions

Our study demonstrates that hidden blood loss exists prior to surgery and accounts for more than 80% of the total blood loss in patients with intertrochanteric fracture. It also exists postoperatively. To avoid long-term anemia after surgery, orthopedic staff should closely monitor dynamic changes in Hb. Moreover, to achieve a smooth perioperative period, we should reduce the harm caused by hidden blood loss as far as possible by supplementing the patient’s blood circulation and correcting anemia when necessary.