SUMMARY ORDER

UPON DUE CONSIDERATION, IT IS HEREBY ORDERED, ADJUDGED AND DECREED that the judgment of the District Court is hereby AFFIRMED.

Raul and Mercedes Perez appeal from Judge Scheindlin’s adverse judgment in favor of the United States following a bench trial. Appellants contend that the district court committed clear error in finding that negligence committed by the Department of Veterans Affairs Medical Center in New York (“VAMC”) during Raul Perez’s (“Perez”) prostate surgery was not the proximate cause of their injuries and that the district court ignored evidence that established proximate cause by a preponderance of the evidence. We affirm.

Perez was diagnosed with prostate cancer and underwent a radical perineal prostatectomy at VAMC on August 20, 1993. The procedure involved removing the prostate entirely and then reconnecting the urethra to the bladder. During the procedure, a Foley catheter was inserted through the penis to divert urine, and the catheter remained in place for ten days after the operation to allow the anastomosis — the site of reattachment of the urethra to the bladder — to heal properly.

The VAMC medical staff removed Perez’s catheter during the night of August 30. Several hours later, on the morning of August 31, Perez went into urinary retention and the medical staff made more than *50one unsuccessful attempt to reinsert the catheter. A few hours later, VAMC physicians operated on Perez to reinsert the catheter with aid of a cystoscope, which enabled them to visualize the anastomosis. The cystoscopy revealed no damage to the anastomosis and there was no evidence of the leakage of urine at that time. After the August 31 procedure, the physicians ran a charcoal test to check for the presence of a recto-urethral fístula, which is an unnatural communication between the urethra and rectum. The test was negative.

A week later, on September 7, the physicians performed a cystogram that indicated leakage of urine, which indicated that the anastomosis had become damaged. Three days later, on September 10, the physicians performed another charcoal test that again came back negative for the presence of a recto-urethral fistula. On September 13, a cystogram and a voiding cystourethrogram indicated a leaking anastomosis and a stricture (tightening) of the urethra. On September 20, a methylene blue test revealed an anastomotic leak but demonstrated that no recto-urethral fistula was present. Finally, on September 23, a hypaque enema study was conducted and once again confirmed that no recto-urethral fistula existed. On September 28, with no sign of further leakage of the anastomosis, Perez was discharged from VAMC.

On November 30, Perez returned to VAMC complaining of symptoms that indicate a recto-urethral fistula. On December 14, VAMC physicians examined Perez and confirmed that he did indeed have such a condition. As a result of the fistula, the unsuccessful attempts to repair it, and its failure to heal spontaneously, Perez has suffered numerous painful and debilitating symptoms.

On August 12, 1998, Perez and his wife brought this lawsuit under the Federal Tort Claims Act, 28 U.S.C. § 1346 (“FTCA”), alleging various acts of negligence on the part of the VAMC physicians. Applying New York law, which governs this case, see 28 U.S.C. § 1346(b)(1), the district judge found that the removal of the catheter on the tenth postoperative day constituted negligence, but rejected appellants’ other allegations of negligence. The district judge found, however, that appellants failed to prove by a preponderance of the credible evidence that the early removal of the catheter caused the fistula and hence the damages suffered by Mr. and Mrs. Perez. Importantly, the district judge found Perez’s testimony regarding the alleged attempts to reinsert the catheter, and appellants’ expert’s testimony, to be not credible. Although the district judge found that it was not possible to determine, on the trial record, what the precise cause of the fistula was, she found that appellants failed to prove that the early removal of the catheter — the only negligent act — caused the leak. She therefore entered judgment for the United States. This appeal followed.

We review a district judge’s findings of fact for clear error, see Fed. R. Civ.P. 52(a); Anderson v. City of Bessemer City, 470 U.S. 564, 573-74,105 S.Ct. 1504, 84 L.Ed.2d 518 (1985), and overturn such findings only if, “on the entire evidence,” we are “left with the definite and firm conviction that a mistake has been committed,” United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 92 L.Ed. 746 (1948). If there are two reasonable views of the evidence, the district judge’s decision to accept one rather than the other cannot be clearly erroneous, and we may not overturn findings of fact merely because we take a different view of the evidence. See Mobil Shipping & Transp. Co. v. Wonsild Liquid Carriers Ltd., 190 F.3d 64, 67 (2d Cir. *511999). Further, where the district court’s finding “is based on [its] decision to credit the testimony of one of two or more witnesses, each of whom has told a coherent and facially plausible story that is not contradicted by extrinsic evidence, that finding, if not internally inconsistent, can virtually never be clear error.” Anderson, 470 U.S. at 575,105 S.Ct. 1504.

Given that the district judge found, and the government does not now contest, that the timing — although not the manner — of the decatheterization was negligent, and given that the leaking anastomosis concededly played a role in the formation of the fistula, the key issue is whether the timing of the decatheterization caused the persistent leakage. Appellants primarily argue that the leakage was caused by the early decatheterization either because (i) the ensuing urinary retention necessitated the reinsertion attempts, which in turn damaged the anastomosis, or (ii) the ensuing urinary retention itself created pressure on the anastomosis that caused the damage. The district judge concluded, however, that the preponderance of credible evidence did not support the conclusion that the timing of the decatheterization caused the leakage.

The district judge’s findings with respect to the reinsertion attempts were based largely upon her negative assessment of Perez’s credibility regarding his testimony that the attempts to reinsert the catheter were rough. Perez testified that the medical staff attempted to reinsert the catheter immediately after removing it on the night of August 30. The district judge herself questioned Perez about the sequence of events:

Perez: They pull [the catheter] out, and then they are trying to put it back.

The Court: Right away?

Perez: Immediately.

The Court: Immediately?

Perez: The same one. They had the other ready. They put a new one.

The Court: Right away, in a minute?

Perez: Immediately. Immediately they pull it out they tried to put another one.

The Court: The minute they pulled it out they tried to put another one?

Perez: Yes.

She ultimately found that the reinsertion attempts were made only because Perez went into urinary retention, which occurred only 6-8 hours after the catheter was removed. The district judge permissibly assessed Perez’s credibility based upon the whole of his testimony and disbelieved his testimony that the reinsertion attempts were rough or painful. Her additional findings that no damage to the anastomosis was visible during the cystoscopy and that Perez was not given any medication stronger than Tylenol, accord with this determination.

With regard to appellants’ theory that pressure on the anastomosis resulting from urinary retention caused damage to the anastomosis, no direct evidence to that effect was adduced at trial. Indeed, the only relevant direct evidence — the August 31 cystoscopy belied such a conclusion. Appellants’ strongest argument is therefore based upon circumstantial evidence: an inference drawn from the temporal connection between the decatheterization and the subsequent anastomotic leaking. However, we cannot agree with appellants that the district judge misstated the record when she found that the August 31 cystoscope indicated that the anastomosis was tight and not leaking. While it was uncontroverted at trial that a cystoscope might not detect smaller holes in the anastomosis or urethra, it also was undisputed that: (i) the cystoscope would have picked up tears in the suture line, had they exist*52ed; and (ii) there was no evidence of leakage during the August 31 procedure.

We note as well that the testimony of Dr. Brodherson, appellants’ expert, as to the ability of the cystoscope to detect small holes in the anastomosis was given in the context of speculating that the blind recatheterization attempts had caused the leakage — a possibility that the district judge had discounted because she disbelieved Perez’s testimony as to the nature of the reinsertion attempts. Further, we cannot rely on Dr. Brodherson’s testimony as a basis to overturn the district judge’s findings of fact, given that she found Dr. Brodherson not credible with respect to several aspects of his testimony, including: (i) the propriety of attempting recatheterization without the aid of a visualization technique such as a cystoscopy; (ii) the propriety of using a cystoscope rather than a cystogram during the successful recatheterization procedure; and (in) the need for a suprapubic cystostomy once the fistula was diagnosed. See, e.g., Mathie v. Fries, 121 F.3d 808, 812 (2d Cir. 1997) (“[Although a trial court’s credibility determinations are not completely immune from appeal, a reviewing court owes particularly strong deference where the district court premises its findings on credibility determinations.” (internal quotation marks, citations, and brackets omitted)). Moreover, even if the cystoscopy results do not undermine appellants’ causation theory, there was no credited direct evidence that supported it, and the district judge’s failure to draw an inference based upon rather thin circumstantial evidence was not clear error.

The district judge disagreed with the theory that the ordering of a charcoal test itself proved that the early decatheterization and subsequent recatheterization caused the leakage, because she found that other reasons also supported the ordering of the test. One such reason, supported by the record, is that in the normal course, more than one in ten perineal prostatectomy patients experiences postoperative rectal injury. But even if we were to assume that the test was performed solely because the physicians were concerned that the events of August 30-31 might have damaged the anastomosis, that still does not provide a sufficient basis to overturn the district judge’s conclusion that appellants failed to adduce sufficient proof in her view as to the cause of the damage.

Given the district judge’s permissible holding that there was inadequate credible evidence to allow her to conclude that the early removal and/or subsequent attempts to reinsert the catheter caused the damage to the anastomosis, the likelihood or unlikelihood of other possible causes is irrelevant. As the district judge correctly stated, the government had “no burden to provide alternative proof of causation; it can simply rely on plaintiffs’ failure of proof.” Perez v. United States, No. 98 CIV 5756, slip op. at 27 (S.D.N.Y. Aug. 11, 1999); see, e.g., Arkin v. Gittleson, 32 F.3d 658, 664 (2d Cir. 1994) (stating that under New York medical malpractice law, plaintiff has burden of proving breach of standard of care and that breach proximately caused injuries).

To overturn the district judge’s rulings would require us to negate credibility findings and to hold that appellants proved causation as a matter of law — ie., that they were entitled to the equivalent of a directed verdict. See Fed.R.Civ.P. 50(a). While we might well question whether as trial judges we would have reached the same conclusions as did the district judge, actually being at a trial provides a perspective quite different from, and usually superior to, reading the cold record. Overturning credibility findings and holding that a factual matter was proven as a *53matter of law by the party bearing the burden of proof are beyond our reach as an appellate court in the present circumstances.

We therefore affirm.