PER CURIAM:
Noel D. Woods appeals from the district court’s judgment, after a bench trial, in favor of the United States on Woods’s medical malpractice claims arising under the Federal Tort Claims Act, 28 U.S.C. §§ 1346(b), 2671 et seq. Woods’s suit alleged that deficiencies in the care he received at Bay Pines VA Medical Center caused him to lose his sight. On appeal, Woods challenges an evidentiary ruling by the district court and also challenges the district court’s factual findings and legal conclusions. After thorough review, we affirm.
I.
The essential facts are these. The plaintiff, Woods, is a veteran of the U.S. Air Force. Since the 1970s, he has suffered from a chronic condition known as Reiter’s syndrome. In November 1998, Woods was diagnosed with avascular necrosis in his left hip. Woods opted to receive a total hip replacement. Dr. John Camblin performed the hip replacement surgery (a left total hip arthroplasty) on April 8, 1999, at Bay Pines VA Medical Center in Bay Pines, Florida. Before the surgery, medical staff gave Woods the option of donating his own blood in advance for use in an autologous transfusion if a transfusion became necessary. Woods opted not to donate blood.
The surgery lasted about two hours; according to Dr. Camblin, “[i]t was a routine operation” with no complications. Woods lost approximately 1000 cc of blood during the procedure. At the end of the procedure, Dr. Camblin installed a Constavac, a device that collects blood from the surgical site and either reinfuses the blood back into the patient’s body or drains it out of the patient’s body. Woods was moved to the Post Anesthesia Care Unit (PACU) and then to the Orthopedic floor.
The parties dispute the amount and pattern of Woods’s postsurgical blood loss. Dr. Camblin testified that Woods lost 530 cc of blood, of which 400 cc was reinfused through the Constavac. This would mean Woods lost a net 130 cc of blood after surgery, which, added to the 1000 cc lost during surgery, would amount to a net blood loss of 1130 cc during and after surgery. Woods claims, however, that his actual blood loss had to have been much greater.
In the few days after surgery, Woods showed signs of agitation and confusion. On April 12, 1999, Dr. Camblin noted memory loss and ordered a neurological consultation. On April 14, 1999, six days after the surgery, Woods complained of problems with his vision, saying things looked fuzzy and cloudy. A neurologist examined him on the same day; the neurologist noted visual impairments and ordered an optometry consultation. Over the next few days, Woods complained of further vision problems such as spots and occasional visual hallucinations. On April 20, 1999, an ophthalmologist, Dr. Saebert Chamikles, examined Woods and found a small flame-shaped hemorrhage in his left eye. He recorded Woods’s visual acuity as 20/400 in each eye.
On April 30, 1999, Dr. Chamikles examined Woods again and determined that he was legally blind. Woods checked out of the hospital on that day. Woods’s blindness has since been diagnosed as posterior ischemic optic neuropathy (PION), which is a death of optic nerve tissue due to decreased blood flow.
On February 28, 2002, Woods filed suit against the United States in the United States District Court for the Middle District of Florida under the Federal Tort Claims Act (FTCA), 28 U.S.C. §§ 1346(b), 2671 et seq.,
which allows tort claims
against the United States. Woods claimed that deficiencies in the care he received at the Bay Pines VA Medical Center before, during, and after his hip surgery led to his developing ischemic optic neuropathy (ION) due to blood loss associated with the surgery. Woods alleged, among other things, that Bay Pines medical personnel failed to warn him in advance of the potential risk of blindness, failed to take necessary precautions to deal with his blood loss and reduce the risk of blindness, and failed to monitor him properly during and after the surgery.
The district court conducted an exhaustive bench trial over 13 days. After the trial, the district court ordered judgment in favor of the United States, explaining its reasoning in a lengthy opinion. The court noted that the parties had submitted conflicting expert testimony. After weighing factors such as “the education, training, experience and overall credentials” of each witness and “inconsistencies in testimony and possible bias,” the district judge concluded that the expert witnesses who testified on behalf of the United States “exhibited a level of knowledge and experience superior to that of Plaintiffs experts.” Based on the testimony of Dr. Nancy Newman, a professor of ophthalmology at Emory University, the court concluded that Bay Pines medical personnel could not have reasonably foreseen that Woods would develop PION as a result of his hip surgery. Based on the testimony of Dr. Steven Roth, an associate professor of anesthesiology and director of the neuroanesthesia section at the University of Chicago, the court concluded that the Bay Pines medical personnel did not deviate from the required standard of care when they elected to use the Constavac device to reinfuse Woods’s blood instead of performing a blood transfusion. The district court also noted that it was viewing Woods’s own testimony “with skepticism” and had “doubts about Plaintiffs veracity in reporting his symptoms accurately.” Based on the records and Woods’s own testimony, the district court observed that medical providers considered him “manipulative,” that Woods “was vocal about his ability to apply pressure to get what he wants,” and that Woods made “early threats to enlist the aid of his counsel.”
The district court observed that liability in the case turned on application of Florida tort law. Under Florida law, a claim for medical malpractice requires the plaintiff to show a breach of the standard of care, damages, and a causal relationship between the breach and the damages. The court found, based on testimony presented by both sides, that Woods suffers from posterior ischemic optic neuropathy, as opposed to some other form of ischemic optic neuropathy affecting a different region of the optic nerve. Neither party has challenged this finding on appeal. The court concluded that the care Woods received at the VA met the requisite standard of care. The court further determined, based in part on Dr. Newman’s testimony, that the causes of PION are not well understood, so Woods could not show that there was a deficiency in the presurgery evaluation that led Woods’s medical providers to miss a warning sign that they should have recognized.
The court also noted that PION is extremely rare and further noted, again pointing to Dr. Newman’s testimony, that
PION is normally associated with lengthy open heart or spinal surgery, not hip replacement surgery of a duration and character comparable to Woods’s surgery. The court, therefore, concluded that the standard of care did not demand that the medical providers warn Woods of the risk of PION. The court also found that the medical providers properly monitored the patient during surgery, and no exigencies during the surgery called for a transfusion or other intervention.
As for Woods’s care after surgery, the district court found that Woods’s medical providers responded in an acceptable manner to his postsurgery symptoms, which included memory loss, agitation, and confusion, and his later complaints of visual disturbances. The court found that the standard of care did not require medical providers to administer a blood transfusion after surgery.
The district court next addressed the question of causation. Citing Dr. Newman’s testimony, the court noted that the causes of PION are not well understood and that it is not possible to predict when PION will occur. Based in part on Dr. Newman’s testimony, the court found that the plaintiff had not established that purported deficiencies in the care he received caused him to develop PION. Because Woods established neither a breach of the standard of care nor causation, the court entered judgment in favor of the United States. Woods timely appealed to this court.
II.
Woods’s first argument is that the district court abused its discretion during the bench trial when it admitted and considered evidence regarding whether Woods had consumed alcohol before his surgery and whether Woods had been arrested on assault charges in 1989.
As for the first matter, in the course of cross-examining Woods, the attorney for the United States asked Woods whether he had left Florida each summer since 1985. Woods confirmed that he had, “to get out of ... the summer weather here because it bothers my arthritis.” The attorney asked whether Woods was in Florida in the summer of 1989 and whether he was arrested in 1989. Woods’s attorney objected. The district court judge allowed the defense attorney to continue with the line of questioning and said she would treat the testimony as a proffer subject to the United States’ later establishing its relevance.
The defense attorney then asked Woods about his “drinking history ... of alcohol prior to April 8th, 1999,” the date of the surgery. Woods answered that “alcohol was not my forte. I enjoyed outdoor sports.” Woods’s attorney again objected, and the district court again said it would allow the defense to proceed with the questioning in the form of a proffer subject to the defendant’s later establishing the relevance of the inquiry. The government attorney also inquired whether Woods had told Dr. Camblin three days after the surgery that he had drunk three or four beers before the surgery. Woods denied having made such a statement to Dr. Camblin. As for the second matter, the defense attorney asked Woods, “[0]n July 6th, 1989, ... were you arrested for aggravated assault with regard to a law enforcement officer?” Woods said, “No. Hell, no.” Woods’s attorney moved to strike, but the judge denied the motion, saying she would consider the testimony as a proffer subject to later decision. The questioning then moved on to other subjects.
The parties later submitted written briefs on the admissibility of evidence regarding alcohol use or the supposed July 1989 arrest. In its written proffer, the
government argued that “both Dr. Camblin as well as Dr. Guskiewicz [an expert witness for the plaintiff] testified that alcohol withdrawal can cause symptoms of agitation and confusion” after surgery. Furthermore, a notation in Woods’s medical records suggested Woods may have used alcohol before the surgery, and other evidence suggested that Woods at least occasionally drank alcohol. This, the government claimed, was a relevant area of inquiry. As for the supposed July 1989 arrest, the government claimed that it was relevant to evaluating Woods’s truthfulness in light of Woods’s testimony that he suffered considerable physical debility and had left Florida every summer since 1985 because of his medical condition.
The district court did not allow the defense to submit an arrest report concerning the supposed July 6, 1989, arrest for attempting to strike a police officer with a bottle of rum. The district court did not indicate whether it had admitted or excluded the proffered testimony from Woods’s cross-examination.
We review decisions as to the admission or exclusion of evidence for abuse of discretion.
Alexander v. Fulton County,
207 F.3d 1303, 1326 (11th Cir.2000);
cf. United States v. Frazier,
387 F.3d 1244, 1258-59 (11th Cir.2004) (en banc). “The application of an abuse-of-discretion review recognizes the range of possible conclusions the trial judge may reach.... [W]hen employing an abuse-of-discretion standard, we must affirm unless we find that the district court has made a clear error of judgment, or has applied the wrong legal standard.”
Frazier,
387 F.3d at 1259.
After thoroughly reviewing this record, we can find no indication that the district court actually admitted Woods’s responses to the questions regarding alcohol use or a July 1989 arrest. The district court judge never said she was admitting the testimony, and, notably, made no mention of either matter in her expansive 29-page opinion. But even if the district court in fact did admit the testimony, Woods has not shown that admission of the evidence somehow amounted to reversible error. There is no indication plaintiff’s substantial rights have been affected in any way. The cross-examination produced nothing aside from Woods’s emphatic denials on both subjects.
Moreover, we are not convinced that admission of Woods’s responses to the questions concerning alcohol would have amounted to an abuse of discretion in any event. Evidence of alcohol use was arguably relevant to Woods’s postsurgery symptoms of agitation and confusion. As for the assault issue, we reiterate that there is no indication the district court ever considered it. It is also worth repeating that this was a bench trial. As the former Fifth Circuit stated in
Gulf States Utilities Co. v. Ecodyne Corp.,
635 F.2d 517, 519 (5th Cir. Unit A Jan.1981), the part of Rule 403 that authorizes exclusion of evidence because of its unfair prejudicial impact “has no logical application to bench trials.... Rule 403 assumes a trial judge is able to discern and weigh the improper inferences that a jury might draw from certain evidence, and then balance those improprieties against probative value and necessity. Certainly, in a bench trial, the same judge can also exclude those improper inferences from his mind in reaching a decision.”
III.
Woods next challenges the basic findings of fact and legal conclusions drawn by the district court.
We review a district court’s findings of fact for clear error. Fed. R. Civ. P. 52(a);
Newmann v. United States,
988 F.2d 1258, 1262 (11th Cir.1991). “A finding is ‘clearly erroneous’ when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed.”
United States v. U.S. Gypsum Co.,
333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948). We review conclusions of law de novo.
EEOC v. Joe’s Stone Crab, Inc.,
220 F.3d 1263, 1273 (11th Cir. 2000).
As the district court noted, liability in a suit under the Federal Tort Claims Act turns on whether a private party in like circumstances would be hable under the prevailing law. 28 U.S.C. §§ 1346(b)(1), 2674;
Pate v. Oakwood Mobile Homes, Inc.,
374 F.3d 1081, 1083-84 (11th Cir. 2004) (“The United States can only be found liable if a comparable private party would likewise be liable under [state] law.”). In this case, Florida law provides the rules for determining liability.
Under Florida law, “[t]o prevail in a medical malpractice case a plaintiff must establish the following: the standard of care owed by the defendant, the defendant’s breach of the standard of care, and that said breach proximately caused the damages claimed.”
Gooding v. Univ. Hosp. Bldg., Inc.,
445 So.2d 1015, 1018 (Fla.1984);
see also Torres v. Sullivan,
903 So.2d 1064, 1067 (Fla.Dist.Ct.App.2005). The Florida statute defines the standard of care in medical malpractice claims as “that level of care, skill, and treatment which, in light of all relevant surrounding circumstances, is recognized as acceptable and appropriate by reasonably prudent similar health care providers.” Fla. Stat. § 766.102(1) (1999) (amended 2003).
First, Woods says the district court reversibly erred when it made a factual finding that the amount of Woods’s blood loss in the course of his surgery and hospitalization was about 1130 cc. In the first place, it is not absolutely clear that the district court made a finding on the amount of Woods’s total blood loss. Although the court mentioned that Dr. Camblin placed Woods’s total blood loss at 1130 cc, it noted that that figure was disputed, and when the court entered its own findings, all that it said was that it accepted “Dr. Camblin’s testimony that Plaintiff lost 1000 cc’s during surgery,
and the rest afterward”
(emphasis added).
Woods argues, nevertheless, that the estimate of 1130 cc of blood loss contradicted a notation in his medical records made by Chris Kelly, a student training to be a physician’s assistant. That notation, made after 400 cc of blood had already been drained from and reinfused into Woods’s body, indicated, “Constavac drain in place with 530cc total blood loss.” Woods claims this meant that his cumulative blood loss after surgery was 930 cc — the sum of 400 cc and 530 cc. But Kelly himself testified at trial that the notation of 530 cc would have been a cumulative figure that included the 400 cc that was reinfused into the patient’s body. Plainly, it was not clear error for the district court to accept Kelly’s interpretation of the notation that he himself created.
Woods’s next contention is that Dr. Camblin stated in pretrial deposition testimony that Woods’s blood loss was 1530 cc, and he adopted a new and substantially different estimate of 1130 cc of blood loss at trial. Woods’s attorney was aware of this discrepancy and in fact confronted Dr. Camblin about it at trial. Dr. Camblin explained, however, that his determination had simply changed after further review of the medical records, including Kelly’s notation. It was not clearly erroneous for the district court to accept Dr. Camblin’s ex
planation and favor his revised estimate of blood loss.
Woods also argues that according to guidelines published by the American Society of Anesthesiologists (ASA), his hemoglobin and hematocrit levels after surgery indicated levels of blood loss much higher than 1130 cc. Hemoglobin (Hb) is an oxygen-carrying protein found in red blood cells; hematocrit (HCT) is a measure of the blood’s red cell content. Woods cites an ASA report titled Practice Guidelines for Blood Component Therapy. The report states that “[t]he transfusion of one unit of whole blood or RBCs [red blood cells] increases the hematocrit by approximately 3%, or the hemoglobin concentration by 1 g/dL, in a 70-kg nonbleeding adult.” Pl.’s Ex. 182-46 at 6. Based on this, Woods calculates that his hemoglobin and hematocrit levels after surgery indicate blood loss much higher than 1130 cc. But the ASA report only purports to describe the effect of a
transfusion
of
one unit
of whole blood or red blood cells on hematocrit and hemoglobin levels in a typical
nonbleeding
adult. Nothing in the ASA report suggests that the same rules of thumb can be used as Woods proposes, to accurately determine the volume of blood lost by a bleeding adult who has lost more than one unit of blood. This argument does not show that it was clearly erroneous for the trial court to find that Woods lost approximately 1130 cc of blood.
Woods’s fourth claim relies on two medical journal articles that describe methods for calculating blood loss based on hematocrit levels. But the articles only present these methods as ways to produce estimates of blood loss for purposes of evaluating aggregate outcomes in multiple patients, for purposes such as “comparisons between surgical techniques, between surgeons, and between different institutions,” Pl.’s Ex. 182-45 at 1072. Neither article suggests that the methods described can or should be used to determine the amount of blood lost by an individual patient. Indeed, one of the articles evaluated its proposed method by calculating blood loss for 250 surgery patients and then comparing the results against anesthesiologists’ estimates of blood loss in the same cases. The article determined that there was not a strong correlation between the calculated values and the anesthesiologist-estimated values, suggesting that the method would not be useful for determining blood loss in individual cases.
See
PL’s Ex. 182-45 at 1072,1073 Fig. 4.
Moreover, both the government’s expert witness, Dr. Roth, and the plaintiffs expert witness, Dr. Robert Guskiewicz, an associate professor of neurosurgery and anesthesiology at the University of Florida, testified that practicing anesthesiologists do not typically rely on the methods described in the articles as ways to accurately determine the amount of an individual patient’s blood loss.
Woods’s fifth contention is that a third medical journal article says that a decline in hematocrit comparable to what Woods experienced would only occur if blood loss were greater than 2000 cc. Woods overstates the definitiveness of the article. The cited portion of this article only describes conclusions drawn from a mathematical model of autologous blood donation.
See
Def.’s Ex. 66 at 1619. Like the two articles discussed above, the third article does not suggest that a change in hematocrit can be used by itself to draw reliable conclusions about the amount of blood lost by an individual real-world patient. At all events, the district court was not obliged to credit it.
Next, Woods argues that his medical records indicate symptoms consistent with blood loss greater than 1130 cc. These symptoms included anxiety, restlessness, hypotension, altered mental states, low
urine output, and depressed mental state. But the fact that Woods exhibited symptoms that could be explained by high levels of blood loss does not compel the conclusion that high blood loss in fact caused those symptoms. The defense presented ample evidence showing that the symptoms could have been caused by other factors. A defense expert, Dr. Roth, testified that drugs, pain, and the hospital environment can cause mood and behavioral disturbances, and the plaintiffs expert witness, Dr. Guskiewicz, agreed that agitation can be caused by any of “several” factors, including reaction to medication and alcohol withdrawal. Furthermore, even assuming the symptoms were in fact caused by blood loss, the presence of the symptoms does not demonstrate that the amount of Woods’s blood loss was higher than the 1130 cc amount stated in Dr. Camblin’s testimony. None of the evidence in the case compels the conclusion that Woods’s sensitivity to blood loss exactly matches that of some hypothetical textbook patient.
Next, Woods notes that his expert witness, Dr. Guskiewicz, testified that hip surgery patients often experience extensive postsurgical blood loss. However, Dr. Guskiewicz did not say in the cited portions of his testimony that total blood loss is invariably greater than 1130 cc. In any event, there is nothing definitive about Dr. Guskiewicz’s testimony; other evidence, including Woods’s medical records and Dr. Camblin’s testimony based on those records, also justified a finding that Woods’s blood loss was 1130 cc. Woods also cites to a medical textbook titled
Clinical Anesthesia Practice,
but that textbook was not entered into evidence at trial, so we cannot examine it on appeal.
See Selman v. Cobb County Sch. Dist.,
449 F.3d 1320, 1332 (11th Cir.2006) (“In deciding issues on appeal we consider only evidence that was part of the record before the district court.”);
Shahar v. Bowers,
120 F.3d 211, 212 n. 1 (11th Cir.1997) (en banc).
Woods also cites testimony by Dr. Camblin that hip surgery patients can experience extensive blood loss for up to 24 to 48 hours after surgery. But Dr. Camblin only stated that “there are some people who will bleed beyond” the first six hours, and up to 48 hours. He did not testify that hip surgery patients invariably, or even routinely, experience this level and pattern of blood loss. The vast majority of the bleeding, Dr. Camblin stated, occurs within the first six hours. In short, Dr. Camblin’s testimony does not demonstrate that the district court’s findings were clearly erroneous.
Woods’s final argument regarding the amount of blood loss is that it could not have been as low as 1130 cc because PION is normally associated with blood loss of 2000 cc or greater. But even if PION is usually associated with blood loss at that level, that would not conclusively establish that Woods’s blood loss in fact was in that range. Woods has not demonstrated that his case must have been a typical or average case, and he has not shown that PION is restricted to cases where blood loss exceeds 2000 cc. Indeed, each of the three medical journal articles that Woods has cited in support of his position
identifies one or more cases in which PION or another visual system injury occurred without recorded blood loss greater than 2000 cc.
See
Pl.’s Ex. 169-D at 926; Pl.’s Ex. 182(27) at 386 tbl. 17-2; Pl.’s Ex. 169-H at 2 & tbl. 1. So even Woods’s own evidence suggests that it was reasonable for the district court to conclude that Woods’s blood loss was 1130 cc.
In short, even assuming that the district court in fact made and relied on a factual finding that Woods’s blood loss during and after surgery was approximately 1130 cc, Woods has failed to demonstrate clear error.
IV.
Woods also claims that the district court erred in finding that he had not demonstrated a breach of the required standard of care. Much of Woods’s argument as to the standard of care relies on his earlier assertion that his blood loss had to have been greater than the 1130 cc amount stated in Dr. Camblin’s testimony. But as we explained, it was not clearly erroneous for the district court to accept Dr. Camblin’s estimate. Thus, Woods’s argument again must fail to the extent that it hinges on the assumption that Woods lost more than 1130 cc of blood.
Woods raises three more points that do not depend on the amount of blood loss. First, he says that his expert witness, Dr. Guskiewicz, testified that Woods should have received a transfusion even if the amount of blood loss was as little as 1130 cc. But a defense expert witness, Dr. Roth, emphatically stated in his testimony that the acceptable standard of care did not demand a transfusion, and that Woods’s treatment appeared to have met the standard of care. Again, it was not clear error for the district court to favor Dr. Roth’s testimony over Dr. Guskiewicz’s testimony.
Second, Woods argues that the criteria Dr. Camblin and an anesthesiologist, Dr. Selim Elzayat, used to determine whether a transfusion was necessary were inconsistent with ASA guidelines. Woods again relies on the ASA report titled Practice Guidelines for Blood Component Therapy. The report states that medical providers should not rely exclusively on a “transfusion trigger,” that is, “an absolute hemoglobin or hematocrit value,” or on changes in vital signs. According to the guidelines, medical providers should consider additional factors such as “the patient’s cardiopulmonary reserve ..., the rate and magnitude of blood loss (actual and anticipated), oxygen consumption ..., and atherosclerotic disease.” Pl.’s Ex. 182-46 at 7-8. However, Woods did not show that Dr. Camblin and Dr. Elzayat ignored any factors that would have indicated that he needed a transfusion. More importantly, Dr. Roth testified that no events during Woods’s surgery or recovery called for a transfusion under ASA guidelines or the current standard of care at the time. Dr. Roth explained how he reached that conclusion through a detailed analysis of what the medical records indicated about Woods’s cardiopulmonary health before the surgery and his hemoglobin, blood loss, vital signs, indicators of blood volume such as urine output, and pulse oximeter readings. Again, it was not clearly erroneous for the district court to choose to accept Dr. Roth’s opinion.
Finally, Woods argues that the standard of care required Dr. Camblin to obtain predonated autologous blood in light of the amount of blood loss typical in the surgeries he performed. However, as Woods admits, Dr. Roth testified that use of the Constavac device was an acceptable alternative to the use of predonated autologous blood. Again, it was not clear error for the district court to favor Dr. Roth’s opinion over the opinion of Woods’s expert witness, Dr. Guskiewicz.
In short, Woods has faded to demonstrate clear error in the district court’s finding that Woods’s treatment met the standard of care established by Florida law. Because Woods failed to show a breach of the standard of care, he failed to establish one of the necessary elements for medical malpractice under Florida law,
and he cannot recover from the United States under the Federal Tort Claims Act. Thus, we have no occasion to address Woods’s three remaining assertions: that the district court committed clear error in determining that Woods failed to prove the causal element, that the district court applied an erroneous definition of causation derived from the testimony of Dr. Newman, and that the case should be retried by a different trial judge.
AFFIRMED.