Abstract

Background:Preoperative anemia may affect postoperative mortality and morbidity following elective cranial operations.

Methods:The American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database was used to identify elective cranial neurosurgical cases (2006-2012). Morbidity was defined as wound infection, systemic infection, cardiac, respiratory, renal, neurologic, and thromboembolic events, and unplanned returns to the operating room. For 30-day postoperative mortality and morbidity, adjusted odds ratios (ORs) were estimated with multivariable logistic regression.

Results:Of 8015 patients who underwent elective cranial neurosurgery, 1710 patients (21.4%) were anemic. Anemic patients had an increased 30-day mortality of 4.1% versus 1.3% in non-anemic patients (P P

Conclusions:Preoperative anemia in elective cranial neurosurgery was independently associated with an increased risk of 30-day postoperative mortality and morbidity when compared to non-anemic patients. A hematocrit level below 33% (Hgb 11 g/dl) was associated with a significant increase in postoperative morbidity.

Keywords: Anemia, cranial, hematocrit, hemoglobin, National Surgical Quality Improvement Program, neurosurgery

INTRODUCTION

Anemia has been shown to be an independent risk factor of postoperative mortality and morbidity for several surgical procedures.[ 4 6 9 28 32 47 ] Anemia is of particular concern in clinical neurosurgery due to the substantial metabolic demand of neural tissue. Compensatory mechanisms can alleviate some effects of a low hematocrit, such as autoregulatory changes in cerebral blood flow[ 18 19 26 ] and greater oxygen extraction ratios.[ 26 ] However, an optimal hematocrit value or range has yet to be established.[ 42 ]

The effect of anemia on overall short-term mortality and morbidity following cranial surgery remains uncertain. Preoperative anemia can exacerbate the clinical manifestations of significant intraoperative blood loss, which may range from 50 to 4300 ml[ 48 ] in cranial surgery.[ 26 40 ] With over 574,000 cranial neurosurgical procedures performed in the United States each year,[ 37 ] it is important to assess preoperative anemia as a risk factor for morbidity and mortality. Currently in neurosurgery, anemia is associated with worse outcomes in emergent cases for aneurysmal subarachnoid hemorrhage (SAH);[ 26 ] however, no similar association has been found in elective neurosurgical cases.

The purpose of this study was to determine the effect of preoperative anemia on overall short-term mortality and morbidity in patients undergoing elective cranial surgery. Also, tolerable hematocrit levels and postoperative morbidity profiles were explored for various neurosurgical procedures.

METHODS

Data source

The American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database was used to identify patients undergoing common cranial neurosurgical procedures between 2006 and 2012. NSQIP is a nationally validated, risk-adjusted, prospectively collected database, which provides perioperative prognostic variables.[ 15 ] Over 250 hospitals, both academic and non-academic, participate in NSQIP, with over 2,000,000 patients collected to date. NSQIP has accumulated over 40,000 neurosurgical procedures to date.

Patients

We identified patients undergoing common elective cranial surgeries from 2006 to 2012 using American Medical Association Current Procedural Terminology (CPT) codes. All emergent procedures were excluded. The following CPT codes were included: 61450, 61458, 61460, 61500, 61512, 61519, 61600, 61510, 61545, 61546, 61575, 61576, 61583, 61584, 61680, 61682, 61684, 61690, 61697, 61698, 61700, and 61702. The CPT code descriptions can be found in Table 1 . The six craniotomy or craniectomy procedural categories included in the study are: Pain (decompression of sensory root ganglion or cranial nerves), removal of extra-axial tumor or lesion, removal of intra-axial tumor, excision of sellar/parasellar tumors by craniofacial or orbitocranial approach, arteriovenous malformation (AVM), and aneurysm.

Table 1

Descriptions of CPT codes for cranial neurosurgical procedures

Definition of primary outcome and anemia

The primary outcome was 30-day overall postoperative mortality and morbidity, which was an aggregation of all available complications in the NSQIP database. These complications include wound infection, systemic infection, cardiac, respiratory, renal, neurologic, thromboembolic events, and unplanned returns to the operating room, as seen in Table 2 . Anemia's influence on the primary outcome was assessed. According to the definitions from the World Health Organization,[ 22 ] anemia was defined as a preoperative hematocrit <36% [hemoglobin (Hgb) <11 g/dl] for women and <39% (Hgb <13 g/dl) for men for dichotomous analysis and <39% (Hgb <13 g/dl) for categorical analysis. Hgb values can be determined by dividing the hematocrit level by 3.

Table 2

General characteristics of patients undergoing elective cranial neurosurgery by anemia status

Statistical analyses

The study population was described using summary statistics. Continuous variables are displayed as their mean and standard deviation [mean (SD)], while categorical variables are displayed as the number of patients followed by the corresponding proportion with respect to the exposure group [n (%)]. Comparisons between groups were made using the t-test/Wilcoxon rank-sum test for continuous variables or the Chi-squared test for categorical variables, as appropriate. Results were considered significant if the observed P > 0.05.

For 30-day postoperative mortality and morbidity occurrence, adjusted odds ratios (ORs) were estimated using a multivariable logistic regression model for anemic versus non-anemic patients. Modeling was done using a forward stepwise approach, and the model was adjusted extensively for both clinically and statistically relevant confounders [Tables 3 and 4 ]. That is, preoperative predictors in NSQIP were tested in the univariable regression model. Statistically significant variables in the simple logistical regression were then included in the multiple logistical regression. In total, 17 variables were adjusted for in the multiple logistical regression model: Age, sex, body mass index (BMI), smoking status, operation year, work relative value units, wound classification, current wound infection, transfusion <72 h prior to surgery, cardiovascular, neurological, respiratory, and renal comorbidity, diabetic status, steroid use for chronic condition, length of operation, and history of previous operation within 30 days of the surgery. Lastly, preoperative predictors in NSQIP that contained over 10% missing values were not included in the multivariable regression analysis; however, these variables were not significant in the univariable logistical regression. Thus, these dropped variables would not represent unmeasured confounders in the multivariable results.

Table 3

30-day mortality and morbidity complications of patients undergoing elective cranial neurosurgery by anemia status

Table 4

30-day mortality and morbidity complications of patients undergoing elective cranial neurosurgery by procedure

Data management and statistical analyses were done with STATA/SE 12. In accordance with Johns Hopkins guidelines (which follow the US Code of Federal Regulations for the Protection of Human Subjects), Institutional Review Board approval was not needed or sought for the present study. Data were collected as an ACS-NSQIP quality assurance endeavor and only de-identified data were received.

RESULTS

This study analyzed 8015 prospectively collected patients undergoing elective cranial neurosurgery from 2006 to 2012 in the NSQIP databases. Perioperative prognostic factors including age, sex, BMI, race, smoking status, diabetes, alcohol intake, steroid use, operative time, reoperation, and composite system morbidity variables have been included in Table 5 . The mean age of patients was 56.1 years and the average BMI was 28.6. Of the total population, 1710 patients (21.4%) were anemic with a preoperative hematocrit below 39 for men and below 36 for women. Hematocrit values were collected on an average of 5.3 ± 8.6 days before surgery. The patients underwent six main procedures (craniotomy or craniectomy) for pain (5.1%), extra-axial tumor (25.8%), intra-axial tumor (57.2%), sellar tumor (2.7%), AVM (2.5%), and aneurysm (6.7%).

Table 5

Logistic models for 30-day mortality and morbidity complications for all patients undergoing elective cranial neurosurgery

Medical comorbidities

In general, anemic and non-anemic cohorts had similar ages, BMIs, smoking status, operation times, and alcohol intake within the previous 2 weeks. Anemic patients were more likely to be African American, diabetic, on steroids for a chronic condition, and have had an operation within the previous 30 days [ Table 5 ]. When stratified by procedure type, patients had the same preoperative renal and hepatobiliary morbidity. Patients undergoing craniotomy for pain had the least amount of comorbidities. Patients with aneurysms had the most preoperative cardiovascular, preoperative neurological (beside their main lesion), and respiratory morbidity. Furthermore, patients with an intra-axial tumor had the highest rate of hemato-oncological comorbidity.

Postoperative complications

Anemic patients had an overall 30-day mortality rate of 4.1% versus 1.3% in non-anemic patients (P < 0.001) and an overall 30-day morbidity rate of 25.9% versus 14.1% in non-anemic patients (P < 0.001) [ Table 6 ]. Anemic and non-anemic patients had the same rates of superficial surgical site infection (SSI), deep incisional SSI, organ space SSI, wound dehiscence, pulmonary embolism, progressive renal insufficiency, stroke with neurological deficit, coma greater than 24 h, cardiac arrest requiring cardiopulmonary resuscitation (CPR), and unplanned return to the operating room. Anemic patients had statistically significantly increased rates of pneumonia (2.9% vs. 1.3%), unplanned intubation (3.0% vs. 1.7%), being on the ventilator >48 h (2.6% vs. 0.4%), acute renal failure (0.2% vs. 0.1%), urinary tract infection (4.4% vs. 2.5%), myocardial infarction (0.5% vs. 0.2%), graft/prosthesis failure (0.1% vs. 0.0%), deep vein thrombosis/thrombophlebitis requiring treatment (3.3% vs. 2.1%), sepsis (3.4% vs. 1.4%), and septic shock (1.2% vs. 0.5%) over non-anemic patients, respectively.

Table 6

Logistic models for 30-day complications for patients undergoing elective cranial neurosurgery by procedure

Patients undergoing craniotomy for an aneurysm had statistically significantly higher rates of pneumonia, unplanned intubation, utilization of a ventilator >48 h, urinary tract infection, stroke with neurological deficit, coma >24 h, and septic shock [ Table 2 ]. Sellar tumor patients had the highest rate of postoperative sepsis. The highest rates of unplanned reoperation occurred in patients undergoing craniotomy for sellar tumor at 10.1%. The overall 30-day morbidity rates for all patients undergoing cranial procedures stratified by diagnosis are as follows: 26.5% for aneurysm, 24.7% for sellar tumor, 19.7% for extra-axial tumor, 14.8% for intra-axial tumor, 14.4% for AVM, and 5.6% for pain. The 30-day mortality rates for each diagnosis were: 3.2% for aneurysm, 2.3% for intra-axial tumor, 1.9% for sellar tumor, 1.2% for extra-axial tumor, 0.0% for AVM, and 0.0% for pain.

Logistic regression analyses

Initial regression analyses compared all anemic patients to non-anemic patients undergoing cranial neurosurgery [ Table 3 ]. On univariable regression analysis, anemic patients were found to have statistically significantly higher odds of postoperative mortality (OR = 3.24; 95% CI: 2.34-4.47) and morbidity (OR = 1.70; 95% CI: 1.47-1.97). After a forward selection, stepwise regression model building approach, we adjusted for preoperative general characteristics and comorbidities, which on multivariable regression analysis revealed anemic patients to have statistically significantly higher odds of postoperative morbidity when compared to their non-anemic counterparts (OR = 1.29; 95% CI: 1.03-1.61). Attempts to identify a hematocrit level at which morbidity was increased revealed no differences between anemic and non-anemic patients above the 33% hematocrit level (Hgb 11 g/dl). However, patients with a hematocrit of ≥30 (Hgb 10 g/dl) but less than 33 (Hgb 11 g/dl) had increased odds of morbidity (OR = 2.30; 95% CI: 1.55-3.42). Also, all patients with a hematocrit lower than 30 (Hgb 10 g/dl) had increased odds of morbidity (OR = 1.84; 95% CI: 1.01-3.34). Finally, 30-day mortality for anemic patients was almost threefold higher than for non-anemic patients (4.1% vs. 1.3%; OR = 2.77; 95% CI: 1.65-4.66).

When multivariable regression analyses are stratified by procedure type, each cranial procedure group demonstrated increased odds of postoperative morbidity in anemic patients, except extra-axial tumors [ Table 4 ]. Craniotomy for intra-axial tumor (OR = 2.15; 95% CI: 1.24-3.74), sellar tumor (OR = 9.69; 95% CI: 1.04-90.20), and aneurysm (OR = 3.30; 95% CI: 1.45-7.50) all revealed increased odds of postoperative morbidity at the 30-33% hematocrit level (Hgb 10–11 g/dl). All patients undergoing craniotomy for pain with a hematocrit lower than 33% (Hgb 11 g/dl) experienced morbidity. Moreover, all patients with a hematocrit lower than 30% (Hgb 10 g/dl) in the AVM cohort experienced morbidity. These data are consistent with the overall anemic cohort demonstrating increased odds of morbidity below the 33% (Hgb 11 g/dl) hematocrit level.

DISCUSSION

While anemia has been shown to be a risk factor in other surgical fields,[ 4 6 9 28 32 47 ] the association has yet to be elucidated in neurosurgery. Anemia is generally regarded as an indicator of overall patient health status; however, it may also independently predispose patients to worse outcomes.[ 34 ] This study provides evidence that anemia is an independent risk factor for increased postoperative mortality and morbidity events following elective neurosurgical procedures. In addition, a specific level of anemia corresponding to a hematocrit below 33% (Hgb 11 g/dl) was shown to be associated with increased morbidity. When procedures were stratified into craniotomy for pain, extra-axial lesion or tumor, intra-axial tumor, sellar tumor, AVM, and aneurysm, all procedures except those for extra-axial tumor and aneurysm had increased odds of morbidity if patients were anemic. Anemia may be both a primary contributor to post-operative complications as well as a marker for important co-morbidities.

Interpretation of results

Patients undergoing cranial neurosurgery generally have significant underlying pathophysiologic disease, which anemia may indicate.[ 25 ] This study demonstrated anemic patients to be at almost three times higher odds of mortality than non-anemic patients (4.1% vs. 1.3%) and to have almost two times the rate of postoperative complications as non-anemic patients (25.9% vs. 14.1%). Interestingly, any type of SSI was the same between the groups. However, the differences that were noteworthy consisted of increased rates of unplanned intubation, renal and respiratory complications, myocardial infarction, deep vein thrombosis, and sepsis with and without shock. These complications are severe and suggest anemia's association with medical adverse events.

Also, patients undergoing elective craniotomy for aneurysm had the highest rate of adverse events at 26.5%. Aneurysm patients had the highest rates of poor neurologic prognostic markers, such as utilization of a ventilator >48 h, stroke with neurological deficit, and coma >24 h. These data suggest patients with underlying vascular pathology might be at increased risk of postoperative mortality and morbidity compared to those with other neuropathology.

Anemia in neurosurgery

The results of our study differ markedly from those of Alan et al., who studied a cohort of 6576 patients undergoing cranial neurosurgery, of whom 28.4% had mild anemia and 2.7% had either moderate or severe anemia.[ 2 ] Patients with mild anemia and moderate to severe anemia had increased odds (OR: 1.5 and 1.8) of having a prolonged length of stay (LOS), respectively. Craniotomy for malignant brain tumor surgery was analyzed as a subgroup of patients with the conclusion that anemia increased the LOS, but did not increase complication rates or mortality. They concluded that while anemia does not increase the risk of adverse outcomes, it does extend the LOS, thereby increasing resource utilization and costs. These findings differ from our conclusion for various reasons. First, our study population was inclusive of more types of patients undergoing elective cranial neurosurgery for various tumors, vascular anomalies, and nerve compressions. Second, our study has greater statistical power as we analyzed an additional 1500 patients relative to the study of Alan et al. Finally, we defined anemia according to the WHO definitions, classifying women with a hematocrit less than 36% and men with a hematocrit lower than 39% as anemic. Alan et al. defined patients as anemic if they had a hematocrit percentage of less than 38.

A systematic review by Le Roux demonstrated that preoperative anemia is associated with adverse outcomes in SAH patients.[ 26 ] They also concluded that anemia may lead to or exacerbate delayed ischemia in cerebral tissue. Other groups have shown an increase in severe disability, delayed infarction, and even death following SAH in anemic patients.[ 24 ] Other rare neurologic bleeding episodes can happen in the presence of pathologic anemia, such as in sickle cell anemia, hemorrhagic infectious diseases, coagulopathies, dural AVMs, and hemorrhagic tumors.[ 3 33 ] Anemia resulting from aneurysmal SAH increases the risk of symptomatic vasospasm[ 38 ] and is associated with poor 3-month outcomes.[ 44 ]

Anemia has also been shown to be a risk factor for postoperative complications in other neurosurgical subspecialties, including spine. Anemia is associated with postoperative ileus in lumbar fusion,[ 12 ] an increased LOS in posterior lumbar spine surgery,[ 2 17 ] postoperative delirium after lumbar surgery,[ 13 ] postoperative ischemic optic neuropathy,[ 7 ] and overall postoperative morbidity in elective spine surgery.[ 2 ] Anemia also increased the rate of SSIs following spine surgery.[ 1 ] In addition, intraoperative acute blood loss anemia has been associated with perioperative cardiac events following cervical[ 14 ] and lumbar spine surgery.[ 11 ]

Anemic correction

Our study demonstrates preoperative anemia is associated with postoperative mortality and morbidity, highlighting the question of whether the anemia should be treated prior to elective cranial surgery. Anemia may be treated with iron supplementation, erythropoietin, or transfusion of packed red blood cells, among other rare options.[ 29 ] The benefits of correcting low hemoglobin levels must be weighed against the risk of these interventions, especially transfusion therapy, which is associated with infectious agent transmission, pneumonia, impaired pulmonary function, extended ventilator support, acute lung injury, acute respiratory distress syndrome, systemic inflammatory response syndrome, organ failure, transfusion reactions, and a prolonged LOS.[ 5 8 10 16 20 30 31 35 36 39 41 ]

Guidelines for anemia correction have yet to be established within neurosurgery, but laboratory studies of both animals and healthy human participants show a hemoglobin below 7 g/dl impairs brain function and below 10 g/dl hinders recovery following traumatic brain injury.[ 7 42 ] New guidelines for acute upper gastrointestinal bleeding have suggested a more restrictive transfusion strategy for survival benefit,[ 43 ] and such a restrictive transfusion strategy is currently favored in general critical care medicine surrounding acute brain injury.[ 29 ] Due to the association of increased postoperative mortality, morbidity, and increased LOS[ 2 ] of anemic patients after elective cranial surgery, future prospective studies are needed in order to determine the appropriate management of preoperative anemia.

Multiple basic science experiments and clinical studies have identified a hematocrit of 30-35% as a functional threshold for oxygen carrying capacity. Hint was the first to describe settings in which blood loss, unless severe, should not be replaced because a 30% hematocrit was determined to have the highest oxygen transporting capacity.[ 21 ] Wood et al. studied the effects of hypervolemic hemodilution with autologous plasma in dogs, concluding that hypervolemic hemodilution decreases cerebral vascular resistance, possibly due to the secondary changes on blood viscosity, with an additional increase in intracranial pressure.[ 46 ] The study supports the hypothesis that decreased blood viscosity through hemodilution can account for the direct relationship observed between cerebral blood flow and cardiac output. Lee et al. sought to determine which hematocrit level was optimal post-hemodilution for maximum oxygen delivery to the ischemic neural tissue.[ 27 ] The mean infarction volume was determined after isovolemic hemodilution at hematocrit levels of 25%, 30%, 35%, 40%, and in a control group. The infarction size had the lowest average volume at the 30% hematocrit level and was significantly lower than the control group, indicating a hematocrit of 30% is optimal for brain protection. Studies of hemorheology have determined factors contributing to blood viscosity include hematocrit, erythrocyte aggregation and flexibility, platelet accumulation, and plasma constituents.[ 23 45 ] While studies suggest that the optimal rheological performance and brain oxygenation is obtained with a hematocrit of approximately 30%, prospective studies are needed to identify the most appropriate trigger levels for transfusion.

Strengths and limitations

Strengths to this study include the use of a large, prospectively collected database, extensive adjustment for confounding variables following univariable analysis, and the diversity of academic and nonacademic international hospital data. Also, NSQIP is collected following yearly quality checks in a standardized fashion with a dedicated surgical nurse at each institution.[ 15 ] There are limitations to this study. Causation cannot be established due to the observational nature of prospective data of this study. Also, the NSQIP database only provides having up to 30-day postoperative outcome data, preventing extrapolation of conclusions beyond this window period. Only hematocrit data, not hemoglobin, was available in NSQIP. The WHO defines anemia using both definitions and, as such, we used hematocrit levels. Countermeasures taken to correct anemia before surgery cannot be determined in NSQIP. The only countermeasure available is transfusion <72 h prior to surgery, which was controlled for on multivariable analysis. Other factors, such as volume and fasting status, could affect the hematocrit levels as well. Finally, the reason for preoperative anemia cannot be discerned with the database, but potential causes may be inferred such as alcohol use prior to surgery, steroids to control chronic conditions, and cardiovascular, renal, and pulmonary comorbidities. Future studies to investigate the long-term impact of anemia on neurosurgical outcomes are needed.

CONCLUSION

In this analysis, preoperative anemia in elective cranial neurosurgery was independently associated with an increased risk of 30-day overall postoperative mortality and morbidity when compared to non-anemic patients, after adjusting for preoperative characteristics and comorbidities. Multivariable regression analysis revealed anemic patients have a 30-day mortality that is threefold higher than that of non-anemic patients. Also, anemic patients had significantly higher odds of postoperative morbidity when compared to their non-anemic counterparts. A difference in postoperative morbidity odds was demonstrated in patients above and below the 33% hematocrit level (Hgb 11 g/dl), suggesting this level as a threshold for increased morbidity.

The authors present compelling information gleaned from the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database that there is a strong association between preoperative ischemia and preoperative morbidity and mortality. This dataset is a multi-institutional review of 8015 elective craniotomy patients, 1710 of whom were anemic preoperatively by WHO standards. Both 30-day postoperative morbidity and mortality were dramatically statistically significantly increased in anemic patients, controlling for co-diagnoses and co-morbidities by multivariate analysis. A review of statistical methodology is beyond the expertise of this reviewer.

This data analysis and presentation is well ahead of the wave in presenting the potential advantages of the analysis of “big data,” and these important results speak for themselves. Particular sources of morbidity are UTI, generalized sepsis, and pulmonary complications including pneumonia. Further confirmation by distinct datasets, and prospective research which modulates both the extent and timing of resolution of pre-operative ischemia will be of great interest. Simple scatter plots for data presentation, particularly in an e-journal, might also be both simple and helpful.

That said, this analysis also points out some of the pitfalls of these techniques, and it is these on which I would like to concentrate in this comment. In the body of the paper, the authors are careful to point out the association between anemia and morbidity, and to distinguish between association and causality, but they fall into that very trap, by implication, in their title, by using the multi-definitional word “increases.” The human mind is quintessentially an inference engine, and it is important to take deliberate measures to avoid jumping to conclusions. Of interest, the peak incidence of adverse effects seemed to lie between hematocrits of 30 and 33, more than with anemic hematocrits either above or below this level of optimum blood flow. This emphasizes the extent to which the important associations of preoperative anemia to outcomes are related in clear, but poorly understood, ways.

(Should the authors wish to change the title to “is associated with” from “increases,” I would be happy to modify this point.)

The authors have chosen not to present this study for local IRB approval, from an institution which has a fraught history with research [(see lines 109-113) http://pages.jh.edu/~jhumag/0202web/trials.html]. I believe this is problematic, as considerable recent work has proven that the veil of data anonymization is easily pierced by reverse data analysis (http://www.j-biomed-inform.com/article/S1532-0464(04)00053-X/abstract). Our society has not yet devised a mature and generally accepted response to questions of research privacy. An overcautious approach is, therefore, warranted in this circumstance, though I imply no bad faith on the part of the investigators. The 17 variables of the multiple logistical regression model therefore provide confounding not only of anonymity, but also insofar as they may not be not fully independent, a potential confounding of results. While some see the issue of research consent as overwrought in these circumstances, a failure to respect the obligations of investigators, who have no inherent right to the data mined, is likely ultimately to exacerbate the public distrust of medical science and the practitioners whose first duty must be to the patient, and ultimately further shackle access to scientific evaluation as well.

Finally, it is important to maintain, particularly in “big data” where the quality of input data and data acquisition is less certain, a healthy uncertainty toward the data itself, though this is somewhat mitigated by statistical averaging in larger series. Definitions are not provided of progressive renal insufficiency, acute renal failure, or septic shock; coma >24 h and ventilation >48 h are sometimes deliberate therapeutic strategies and therefore arguably not necessarily morbidities; and the specific definitions of the variables for multivariate analysis are not provided, and have only presumably been standardized and validated inter-institutionally.

Getting “big data” right is very hard, though potentially extraordinarily useful.

References

1. Abdul-Jabbar A, Takemoto S, Weber MH, Hu SS, Mummaneni PV, Deviren V. Surgical site infection in spinal surgery: Description of surgical and patient-based risk factors for postoperative infection using administrative claims data. Spine. 2012. 37: 1340-5

2. Alan N, Seicean A, Seicean S, Neuhauser D, Weil RJ. Impact of preoperative anemia on outcomes in patients undergoing elective cranial surgery. J Neurosurg. 2014. 120: 764-72

3. Babatola BO, Salman YA, Abiola AM, Okezie KO, Oladele AS. Spontaneous epidural haematoma in sickle cell anaemia: Case report and literature review. J Surg Tech Case Rep. 2012. 4: 135-7

4. Beattie WS, Karkouti K, Wijeysundera DN, Tait G. Risk associated with preoperative anemia in noncardiac surgery: A single-center cohort study. Anesthesiology. 2009. 110: 574-81

5. Blajchman MA. Transfusion-associated immunomodulation and universal white cell reduction: Are we putting the cart before the horse?. Transfusion. 1999. 39: 665-70

6. Carson JL, Duff A, Poses RM, Berlin JA, Spence RK, Trout R. Effect of anaemia and cardiovascular disease on surgical mortality and morbidity. Lancet. 1996. 348: 1055-60

7. Chang SH, Miller NR. The incidence of vision loss due to perioperative ischemic optic neuropathy associated with spine surgery: The Johns Hopkins Hospital Experience. Spine. 2005. 30: 1299-302

8. Ciesla DJ, Moore EE, Johnson JL, Burch JM, Cothren CC, Sauaia A. A 12-year prospective study of postinjury multiple organ failure: Has anything changed?. Arch Surg. 2005. 140: 432-8

9. Dunne JR, Malone D, Tracy JK, Gannon C, Napolitano LM. Perioperative anemia: An independent risk factor for infection, mortality, and resource utilization in surgery. J Surg Res. 2002. 102: 237-44

10. Earley AS, Gracias VH, Haut E, Sicoutris CP, Wiebe DJ, Reilly PM. Anemia management program reduces transfusion volumes, incidence of ventilator-associated pneumonia, and cost in trauma patients. J Trauma. 2006. 61: 1-5

11. Fineberg SJ, Ahmadinia K, Patel AA, Oglesby M, Singh K. Incidence and mortality of cardiac events in lumbar spine surgery. Spine. 2013. 38: 1422-9

12. Fineberg SJ, Nandyala SV, Kurd MF, Marquez-Lara A, Noureldin M, Sankaranarayanan S. Incidence and risk factors for postoperative ileus following anterior, posterior, and circumferential lumbar fusion. Spine J. 2014. 14: 1680-5

13. Fineberg SJ, Nandyala SV, Marquez-Lara A, Oglesby M, Patel AA, Singh K. Incidence and risk factors for postoperative delirium after lumbar spine surgery. Spine. 2013. 38: 1790-6

14. Fineberg SJ, Oglesby M, Patel AA, Singh K. Incidence and mortality of perioperative cardiac events in cervical spine surgery. Spine. 2013. 38: 1268-74

15. Fink AS, Campbell DA, Mentzer RM, Henderson WG, Daley J, Bannister J. The National Surgical Quality Improvement Program in non-veterans administration hospitals: Initial demonstration of feasibility. Ann Surg. 2002. 236: 344-53

16. Gong MN, Thompson BT, Williams P, Pothier L, Boyce PD, Christiani DC. Clinical predictors of and mortality in acute respiratory distress syndrome: Potential role of red cell transfusion. Crit Care Med. 2005. 33: 1191-8

17. Gruskay JA, Fu MC, Bohl DD, Webb ML, Grauer JN. Factors affecting length of stay following elective posterior lumbar spine surgery: A multivariate analysis. Spine J. 2013. p.

18. Hare GM. Anaemia and the brain. Curr Opin Anaesthesiol. 2004. 17: 363-9

19. Hare GM, Tsui AK, McLaren AT, Ragoonanan TE, Yu J, Mazer CD. Anemia and cerebral outcomes: Many questions, fewer answers. Anesth Analg. 2008. 107: 1356-70

20. Hebert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med. 1999. 340: 409-17

21. Hint H. The pharmacology of dextran and the physiological background for the clinical use of rheomacrodex and macrodex. Acta Anaesthesiol Belg. 1968. 19: 119-38

22. .editors. Iron Deficiency Anemia Assessment, Prevention and Control A guide for programme managers. Report No.: WHO/NHP/01.3. Report. World Health Organization; 2011. p.

23. Kee DB, Wood JH. Rheology of the cerebral circulation. Neurosurgery. 1984. 15: 125-31

24. Kramer AH, Gurka MJ, Nathan B, Dumont AS, Kassell NF, Bleck TP. Complications associated with anemia and blood transfusion in patients with aneurysmal subarachnoid hemorrhage. Crit Care Med. 2008. 36: 2070-5

25. Kulier A, Levin J, Moser R, Rumpold-Seitlinger G, Tudor IC, Snyder-Ramos SA. Impact of preoperative anemia on outcome in patients undergoing coronary artery bypass graft surgery. Circulation. 2007. 116: 471-9

26. Le Roux PD. Participants in the International Multi-disciplinary Consensus Conference on the Critical Care Management of Subarachnoid Hemorrhage. Anemia and transfusion after subarachnoid hemorrhage. Neurocrit Care. 2011. 15: 342-53

27. Lee SH, Heros RC, Mullan JC, Korosue K. Optimum degree of hemodilution for brain protection in a canine model of focal cerebral ischemia. J Neurosurg. 1994. 80: 469-75

28. Leichtle SW, Mouawad NJ, Lampman R, Singal B, Cleary RK. Does preoperative anemia adversely affect colon and rectal surgery outcomes?. J Am Coll Surg. 2011. 212: 187-94

29. LeRoux P. Haemoglobin management in acute brain injury. Curr Opin Crit Care. 2013. 19: 83-91

30. Malone DL, Dunne J, Tracy JK, Putnam AT, Scalea TM, Napolitano LM. Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma. 2003. 54: 898-905

31. Morisaki H, Sibbald WJ. Tissue oxygen delivery and the microcirculation. Crit Care Clin. 2004. 20: 213-23

32. Myers E, O’Grady P, Dolan AM. The influence of preclinical anaemia on outcome following total hip replacement. Arch Orthop Trauma Surg. 2004. 124: 699-701

33. Ng WH, Yeo TT, Seow WT. Non-traumatic spontaneous acute epidural haematoma - report of two cases and review of the literature. J Clin Neurosci. 2004. 11: 791-3

34. Partridge J, Harari D, Gossage J, Dhesi J. Anaemia in the older surgical patient: A review of prevalence, causes, implications and management. J R Soc Med. 2013. 106: 269-77

35. Ranucci M, Pavesi M, Mazza E, Bertucci C, Frigiola A, Menicanti L. Risk factors for renal dysfunction after coronary surgery: The role of cardiopulmonary bypass technique. Perfusion. 1994. 9: 319-26

36. Refaai MA, Blumberg N. Transfusion immunomodulation from a clinical perspective: An update. Expert Rev Hematol. 2013. 6: 653-63

37. Roberts L. Tracking Neurosurgery: AANS National Neurosurgical Procedural Statistics Report. AANS NEUROSURGEON: AANS NEWS. 2013. 22:

38. Sampson TR, Dhar R, Diringer MN. Factors associated with the development of anemia after subarachnoid hemorrhage. Neurocrit Care. 2010. 12: 4-9

39. Schreiber GB, Busch MP, Kleinman SH, Korelitz JJ. The risk of transfusion-transmitted viral infections. The Retrovirus Epidemiology Donor Study. N Engl J Med. 1996. 334: 1685-90

40. Sena MJ, Rivers RM, Muizelaar JP, Battistella FD, Utter GH. Transfusion practices for acute traumatic brain injury: A survey of physicians at US trauma centers. Intensive Care Med. 2009. 35: 480-8

41. Shorr AF, Duh MS, Kelly KM, Kollef MH. Red blood cell transfusion and ventilator-associated pneumonia: A potential link?. Crit Care Med. 2004. 32: 666-74

42. Utter GH, Shahlaie K, Zwienenberg-Lee M, Muizelaar JP. Anemia in the setting of traumatic brain injury: The arguments for and against liberal transfusion. J Neurotrauma. 2011. 28: 155-65

43. Villanueva C, Colomo A, Bosch A, Concepcion M, Hernandez-Gea V, Aracil C. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013. 368: 11-21

44. Wartenberg KE, Mayer SA. Medical complications after subarachnoid hemorrhage: New strategies for prevention and management. Curr Opin Crit Care. 2006. 12: 78-84

45. Wood JH, Kee DB. Hemorheology of the cerebral circulation in stroke. Stroke. 1985. 16: 765-72

46. Wood JH, Simeone FA, Kron RE, Snyder LL. Experimental hypervolemic hemodilution: Physiological correlations of cortical blood flow, cardiac output, and intracranial pressure with fresh blood viscosity and plasma volume. Neurosurgery. 1984. 14: 709-23

47. Wu WC, Schifftner TL, Henderson WG, Eaton CB, Poses RM, Uttley G. Preoperative hematocrit levels and postoperative outcomes in older patients undergoing noncardiac surgery. JAMA. 2007. 297: 2481-8

48. Zetterling M, Ronne-Engstrom E. High intraoperative blood loss may be a risk factor for postoperative hematoma. J Neurosurg Anesthesiol. 2004. 16: 151-5