This cause presents five issues for our review: (1) Is Dr. Kalb’s opinion “some evidence” for the. award of TTD and medical benefits? (2) Was Garrett required to prove that the aggravation of his arthritic condition caused his disability and need for medical benefits? (3) Did the commission’s confirmation of the DHO’s May 17, 1988 finding that Garrett’s condition was permanent preclude the commission’s continuation of TTD? (4) Did the commission sufficiently explain its reasoning? and (5) Assuming reversal of the writ vacating Garrett’s award, is Chrysler entitled to reimbursement from the Surplus Fund under former R.C. 4123.515 due to the commission’s failure to appeal? For the reasons that follow, we hold that (1) Dr. Kalb’s reports are some evidence to support the commission’s decision; (2) Garrett was required to and did provide evidence to establish a causal connection between his injury and disability; (3) the commission complied with State ex rel. Eaton Corp. v. Lancaster (1988), 40 Ohio St.3d 404, 534 N.E.2d 46, such that the commission’s permanency determination did not preclude subsequent TTD; (4) the commission’s explanation was adequate under State ex rel. Mitchell v. Robbins & Myers, Inc. (1983), 6 Ohio St.3d 481, 6 OBR 531, 453 N.E.2d 721; and (5) Chrysler is not entitled to reimbursement from the Surplus Fund. Accordingly, we reverse.
Some Evidence and Causation
To receive workers’ compensation for conditions developing after an industrial injury, the claimant must show “not only that his injury arose out of and in the course of his employment, but that a direct and proximate causal relationship existed between his accidental injury and his harm or disability.” Fox v. Indus. Comm. (1955), 162 Ohio St. 569, 576, 55 O.O. 472, 475, 125 N.E.2d 1, 5.
The commission relied on the reports of Dr. Kalb and Drs. Purewal and Rosenblatt, both commission specialists, to grant Garrett TTD and authorize surgery. Neither the Purewal nor Rosenblatt reports, however, recognized the aggravation of Garrett’s arthritic condition as allowed by the commission. Thus, the court of appeals correctly concluded that neither report is evidence tying Garrett’s injury to his arthritic disability.
In April 1988, Dr. Kalb reported:
“In summary, it is my opinion [Garrett’s] pre-existing patellofemoral arthritis condition was aggravated by his work-related injury * * *. It is also my opinion at the present time that he has reached maximum medical improvement unless further treatment is carried out. This treatment would consist of patellectomy or knee arthroplasty and would be expected to improve his condition and function *162and possibly allow return to work. His work restrictions after such a procedure, however[,] would involve limitations to avoid frequent knee flexion beyond 90 degrees. This would include restrictions on frequent squatting, stair climbing and ladder climbing activities.
“With his age, muscle strength, and weight, knee arthroplasty would be expected to provide a more favorable lasting result than patellectomy. In direct answer to your question regarding his initial complaints, his symptoms were present on both sides subsequent to his injury. His impairment would be based upon his limitation of knee ROM according to the AMA guidelines for impairment.”
In January 1991, Dr. Kalb recommended replacement of Garrett’s knee and offered this report, apparently to redress inconsistency in two depositions he had provided in February 1990 for the common pleas court proceedings:
“As I mentioned in my deposition, it is unusual that an injury such as that sustained by Mr. Garrett would be expected to produce abrupt, continuous and progressive symptoms of pain within the knee. My opinion regarding the accident resulting in substantial aggravation to his knee condition is based as I mentioned on Mr. Garrett’s history of not having had any problems with his knees of any sort, nor any requirement for medical treatment for his knees prior to the accident * * *. a * * *
“In summary, as I mentioned in my deposition, one would not expect an injury from the side to aggravate or contribute to his patellofemoral arthritic knee condition. However, one must not ignore the patient’s history of having no pain prior to the accident with his knees and having had no medical treatment prior to the accident with his knees. Based upon his history * * * of having [the] onset of symptoms which have become progressive subsequent to the accident[,] [it] is logical to conclude that for whatever reason, his pain did develop immediately subsequent to the accident and therefore it is logical to conclude that the accident certainly played a role in his symptoms even though his radiographic and arthroscopic findings would not likely be expected to be substantially different than prior to the accident. With osteoarthritis or chondromalacia patella, patients are known in many cases to have rather sudden onset of pain which is continuous following a relatively minor traumatic event. •
“In review of prior radiographs on patients such as this, it is clear that the degenerative process had been going on for some time prior to their becoming symptomatic. It is believed that these conditions may become rather abruptly painful following minor trauma or in some cases no trauma, due to the final thin layer of articulate cartilage being finally worn off exposing direct contact with the bone beneath.”
*163Dr. Kalb’s reports represent that Garrett’s injury provoked his arthritic disability; however, Chrysler complains that Dr. Kalb testified differently during his two depositions.2 According to Chrysler and the court of appeals, Dr. Kalb’s deposition testimony conceded his uncertainty about the cause of Garrett’s disability, making his opinion too equivocal to constitute some evidence for the commission’s decision under State ex rel. Owens-Corning Fiberglas Corp. v. Indus. Comm. (1994), 70 Ohio St.3d 263, 265, 638 N.E.2d 565, 567 (commission cannot rely on a repudiated medical opinion or one that is “merely equivocal”).
Chrysler relies on this cross-examination from Dr. Kalb’s first deposition:
“Q. Let me ask you to assume just hypothetically that the events of December 21st, 1985 did aggravate a pre-exsting osteoarthritic condition in Mr. Garrett’s knees. Is there any way you can say to a reasonable degree of medical probability that Mr. Garrett’s knees would not be in the same condition that they are in today, even if the events of December 21st, hadn’t happened? In other words, can you say that he would not have gotten to the point that he is today even if the events of December 21st, 1985 hadn’t happened?
“A. No, I cannot say that.
“Q. And based on the pre-existing congential problems which were significant in Mr. Garrett’s knees, it would not be unfair to assume that he would be in the same condition today that he’s in even if the events of December 21st hadn’t happened?
“A. That is correct.
U ‡
“Q. So what we’ve got is somebody whose having degenerative problems in his knees and from a medical standpoint there’s no way that you can say that his condition is any worse today than it would otherwise have been even if what happened on December 21st hadn’t happened.
“A. That is correct.”
Chrysler also relies on this cross-examination from Dr. Kalb’s deposition taken two weeks later:
“Q. So you would agree, would you not, that based on the conditions that existed in Mr. Garrett’s knees prior to the events of December 21st, 1985 you would have expected to see just over the normal course of time of degeneration an increase in symptoms and discomfort in his knees?
“A. That is correct.
*164“Q. There is no way that you can say today, Doctor, that Mr. Garrett’s condition is any worse than it would have been even if the accident hadn’t happened?
“A. That is correct.”
In response, Garrett insists that Dr. Kalb rehabilitated his testimony during the second deposition and with his January 1991 medical report. Garrett maintains that (1) Dr. Kalb repeatedly attributed Garrett’s disability entirely to the injury-induced aggravation of his arthritis because Garrett had not experienced subjective symptoms of the condition before the accident, and (2) Dr. Kalb’s inconsistency as to the cause of Garrett’s problems was a mistake he later clarified and corrected. Clarification and correction, Garrett argues, resolves equivocation and sustains the evidentiary value of expert opinion. We agree.
The rule that an equivocal medical opinion is “not evidence” on which the commission can rely emerged from State ex rel. Jennings v. Indus. Comm. (1982), 1 Ohio St.3d 101, 1 OBR 135, 438 N.E.2d 420; State ex rel. Paragon v. Indus. Comm. (1983), 5 Ohio St.3d 72, 5 OBR 127, 448 N.E.2d 1372; and State ex rel. Walters v. Indus. Comm. (1985), 20 Ohio St.3d 71, 20 OBR 402, 486 N.E.2d 94. After reviewing this precedent, State ex rel. Eberhardt v. Flxible Corp. (1994), 70 Ohio St.3d 649, 657, 640 N.E.2d 815, 821-822, explained what constitutes equivocation by a medical expert:
“[EJquivocation occurs when a doctor repudiates an earlier opinion, renders contradictory or uncertain opinions, or fails to clarify an ambiguous statement. Ambiguous statements, however, are considered equivocal only while they are unclarified. Paragon, supra. Thus, once clarified, such statements fall outside the boundaries of Jennings, supra, and its progeny.
“Moreover, ambiguous statements are inherently different from those that are repudiated, contradictory or uncertain. Repudiated, contradictory or uncertain statements reveal that the doctor is not sure what he means and, therefore, they are inherently unreliable. Such statements relate to the doctor’s position on a critical issue. Ambiguous statements, however, merely reveal that the doctor did not effectively convey what he meant and, therefore, they are not inherently unreliable. Such statements do not relate to the doctor’s position, but to his communication skills. If we were to hold that clarified statements, because previously ambiguous, are subject to Jennings or to commission rejection, we would effectively allow the commission to put words into a doctor’s mouth or, worse, discount a truly probative opinion. Under such a view, any doctor’s opinion could be disregarded merely because he failed on a single occasion to employ precise terminology. In a word, once an ambiguity, always an ambiguity. This court cannot countenance such an exclusion of probative evidence.”
*165Here, Dr. Kalb offered two different conclusions as to the cause of Garrett’s arthritis; however, the contradiction apparently resulted because of mistake, not uncertainty. His cross-examination testimony was based on the assumption that Garrett had experienced pain and swelling even before his knee injury, which Dr. Kalb later realized was incorrect. On redirect, Dr. Kalb confessed that he had misunderstood his own operative notes and had drawn a conclusion contrary to Garrett’s documented prior medical history. When asked to explain the contradiction, he said:
“Initially when giving the deposition and answering [Chrysler’s counsel’s] questions previously and reviewing this operative note on 3-14-86 I read the sentence: this disorder was made worse by injury.
“And I assumed that this disorder meant [Garrett’s] complaints related to his problem.
“And from that single sentence I was assuming that he had related a history of some prior problem.
“In review of the physical examination and history from that date which was dictated at that time and in more detail, the history and physical examination, that indicates that that is not the case.
“In review of my entire record I have no documentation of [Garrett’s] ever having had a problem or complaining of a problem prior to his injury as noted.
“And so that was inconsistent.
“Either that was an inaccuracy on my part; or by his disorder I meant his congenital disorder, namely chondromalacia patella.
“So either of those two are possibilities; either my * * * disorder sentence of 3-14-86 was not accurate; or more likely, I was referring to this disorder being the congenital problem with his patellofemoral malalignment because that directly conflicts with the dictated history and physical in terms of his having or not having had prior symptoms with his knees.”
And, on further redirect, the following exchange took place:
“Q. Now that you [Dr. Kalb] have had an opportunity to look at the entire document and to clarify the history, does that change your answers to [Chrysler’s counsel’s] questions in reference to [Garrett’s] pre-existing condition?
a * * *
“A. Yes; that is correct.
“Q. And in reference to this document, which history is the reliable history?
“A. The history and physical examinations intended to provide the history. The information dictated on * * * the operative note is designed to give the indications for the procedure.
*166“Q. So they serve two separate purposes?
“A. That is correct.”
Whether we or the court of appeals is persuaded by this explanation is not at issue; ours are not the credibility determinations that count. The commission is the exclusive evaluator of the weight and credibility to be given medical reports of record, and reviewing courts cannot second-guess the commission’s credibility determinations in mandamus. State ex rel. Pass v. C.S.T. Extraction Co. (1996), 74 Ohio St.3d 373, 376, 658 N.E.2d 1055, 1058; State ex rel. Consolidation Coal Co. v. Indus. Comm. (1997), 78 Ohio St.3d 176, 177, 677 N.E.2d 338, 341. Thus, review extends only to whether some evidence exists for the commission’s decision; after that, courts must defer to the commission’s determination.
For these reasons, this court in Eberhardt defended the commission’s reliance on ah ambiguous medical conclusion that was ultimately clarified by the doctor. There we concluded that this was a credibility determination for the commission to make. The same rule must apply when a physician offers contradictory conclusions by mistake, but later acknowledges the contradiction and resolves it to the commission’s satisfaction. In both cases, any uncertainty is destroyed and the reliability of the physician’s statement is rejuvenated.
Accordingly, we hold that where, as here, an inconsistent or contradictory medical conclusion can be attributed to mistake, as compared to lack of knowledge, and testimony or other medical evidence exists from which the commission could conclude that the mistake was resolved, the commission is able to rely on that evidence in granting or denying compensation or benefits. See, e.g., State ex rel. Owens-Corning, supra, 70 Ohio St.3d at 266, 638 N.E.2d at 567 (Douglas, J., dissenting) (where some evidence exists for the commission’s decision, it must not be disturbed in mandamus, notwithstanding “[a]rtful cross-examination”). The court of appeals erred, therefore, in rejecting Dr. Kalb’s testimony as too equivocal.
We also reject Chrysler’s argument that the commission had no evidence upon which to attribute Garrett’s disability completely to the injury-induced aggravation of his arthritic condition. This argument is again based on the theory that Dr. Kalb’s opinion is unreliable, although Chrysler frames it as an attack on a practice purportedly used by the commission — to attribute disability caused by a nonallowed pre-existing condition to the allowed aggravation of that condition.
The commission cannot compensate claimants unless their disability results exclusively from an allowed condition. State ex rel. Wean United, Inc. v. Indus. Comm. (1993), 66 Ohio St.3d 272, 274, 611 N.E.2d 828, 829. Here, Dr. Kalb’s reports and testimony satisfy this standard because he attributed Garrett’s disability entirely to the aggravation of his arthritis, not to the pre-existing condition. Dr. Kalb concluded that while Garrett probably had arthritic changes *167prior to his injury that were detectable by radiological and arthroscopic tests, he had not experienced symptoms that had prevented him from working at Chrysler before his injury, which is the test for temporary total disability. State ex rel. Bowie v. Greater Cleveland Regional Transit Auth. (1996), 75 Ohio St.3d 458, 461, 663 N.E.2d 926, 929. Thus, contrary to Chrysler’s argument, evidence exists from which the commission could find that Garrett had no pre-existing disability attributable to his pre-existing condition. As State ex rel. Kettering Med. Ctr. v. Wallace (1994), 68 Ohio St.3d 588, 589, 629 N.E.2d 444, 446, explained:
“[The employer’s] argument erroneously assumes that evidence of a preexisting condition is prima facie evidence of a pre-existing disability as well. [The employer], however, ignores that claimant worked without any apparent problems prior to her industrial accident. Because her [pre-existing] psychological condition did not affect her ability to work before the accident, [the employer] cannot persuasively argue that the claimant had a pre-existing emotional disability. See Marshall v. Ouachita Hosp. (1980), 269 Ark. 958, 961, 601 S.W.2d 901, 902. (Appellate court upheld claimant’s contention that although he was a polio victim, he had no pre-existing disability in the workers’ compensation sense, since he had been able to work as a lab technician and perform all required tasks for twenty-two years prior to his work-related accident.)” (Emphasis sic.)
Accordingly, we reverse the court of appeals’ finding that the commission abused its discretion by awarding TTD and medical benefits without evidence tying Garrett’s disability to his industrial injury.
Permanency and Subsequent TTD Award
As an alternative basis for affirming the court of appeals’ judgment, see Morgan v. Cincinnati (1986), 25 Ohio St.3d 285, 289-290, 25 OBR 337, 340-341, 496 N.E.2d 468, 472 (where necessary to preserve lower court judgment, R.C. 2505.22 allows appellee to assert assignments of error without filing cross-appeal), Chrysler argues that the commission had no jurisdiction to award Garrett TTD after having found his condition permanent and, effectively, at maximum medical improvement, pursuant to the DHO order of May 17, 1988. We disagree.
As mentioned, in State ex rel. Eaton Corp. v. Lancaster, 40 Ohio St.3d at 407, 534 N.E.2d at 50, we invalidated the commission’s policy, employed initially here when Garrett filed his PTD application, that permitted its hearing officers to continue PTD for “lengthy intervals” between the decision to terminate TTD due to permanency and the decision to grant or deny PTD. But out of concern for claimants’ welfare, we did not immediately terminate TTD for those claimants who were receiving it pursuant to the commission’s defunct procedure. Rather, we directed the commission to implement a new process for assessing PTD eligibility, one that did not compromise the TTD requirement that a claimant’s *168condition remain temporary while this compensation is paid, and to do so “within ninety days of [our] decision, or as soon thereafter as is practicable.” (Emphasis added.) Id. at 408, 534 N.E.2d at 51. Thus, in that case and others, see, e.g., State ex rel. Peabody Coal Co. v. Indus. Comm. (1989), 41 Ohio St.3d 5, 534 N.E.2d 347, reconsidered and modified on other grounds (1989), 44 Ohio St.3d 104, 541 N.E.2d 74, we returned causes for the commission to rectify its having continued TTD despite a prior determination of permanency.
In Garrett’s case, Chrysler appealed the allowance of his aggravated arthritic condition to common pleas court pursuant to former R.C. 4123.519 shortly after we decided Eaton. As this appeal removed the claim from the commission’s jurisdiction, State ex rel. Gatlin v. Yellow Freight Sys., Inc. (1985), 18 Ohio St.3d 246, 249, 18 OBR 302, 305, 480 N.E.2d 487, 490, the commission had little if any meaningful opportunity to comply with our decision. Only after the court of appeals reinstated the commission’s order and invoked the commission’s duty to conduct further consistent proceedings, see former R.C. 4123.519(F) (now R.C. 4123.512[G]) (commission and bureau of workers’ compensation administrator must execute final court judgment as if it were the commission’s decision), did the commission have the power and practicable ability to correct the problem Eaton identified. And the commission made the appropriate correction when it regained jurisdiction — the commission assigned Garrett’s PTD claim to its “Eaton docket,” a procedure that sanctioned expedited review of PTD applications in conjunction with motions to terminate TTD. Cf. State ex rel. Ford Motor Co. v. Indus. Comm. (1992), 65 Ohio St.3d 17, 599 N.E.2d 261 (TTD award after permanency determination invalidated under Eaton because commission had continuous jurisdiction over claim and did not expeditiously comply). We have since endorsed this procedure, State ex rel. Blake v. Indus. Comm. (1992), 65 Ohio St.3d 453, 455, 605 N.E.2d 23, 25; State ex rel. Kinnear Div., Harsco Corp. v. Indus. Comm. (1997), 77 Ohio St.3d 258, 265-266, 673 N.E.2d 1290, 1296, and Chrysler offers neither evidence nor argument to establish that the commission failed to act “as soon as practicable” in making this assignment. Thus, under these circumstances, we consider the commission’s response time to our admonition in Eaton to comply with that decision. Accordingly, we reject Chrysler’s argument that the commission’s prior determination of permanency precluded its further award of TTD to this claimant.
For those claimants who were receiving TTD benefits despite a finding of permanence at the time Eaton was decided, we directed the commission “as soon thereafter as is practicable, to hold hearings to determine the eligibility of these claimants for total disability benefits.” (Emphasis added.) Eaton, 40 Ohio St.3d at 408, 534 N.E.2d at 51. We did not limit the issue at such hearings to a determination of permanent total disability. Instead, it was contemplated that the commission would, at these hearings, determine both the claimant’s continued *169eligibility for TTD and eligibility for PTD. Otherwise, there would have been no need for the court in Eaton to consider, as it did, whether “[i]n each case, the relied-on medical evidence uniformly indicated a permanent condition.” Id. at 411, 534 N.E.2d at 53. Thus, under the preseñt circumstances, there was no need for the commission to premise its continuing TTD on new and changed circumstances.
Eaton notwithstanding, the claimant’s need for surgery, under the facts of this case, constitutes a new and changed circumstance. It is true, as pointed out in the concurring and dissenting opinion, that Dr. Kalb predicted in 1988 that Garrett would benefit from patellectomy, or knee replacement. However, no request for authorization was made at that time, and this issue was neither presented nor decided at the 1988 hearing. It was not until after the 1988 administrative proceedings that Dr. Kalb, in a report dated September 11, 1989, devised a plan for claimant to “apply for Worker’s Compensation [sic ] approval for bilateral knee replacement arthroplasty,” noting “progressive pain in both knees” and that such “knee pain has become intolerable now bilaterally.”
Thus, at the time claimant’s condition was found permanent in 1988, the issue whether claimant would need surgery, and whether such surgery would improve claimant’s condition, was not litigated and, under these facts, could not have been litigated. When the advent of further surgery arose after 1988, there was a new and changed circumstance which justified modification pursuant to R.C. 4123.52. See Stainless Specialty Mfg. Co. v. Indus. Comm. (1985), 144 Ariz. 12, 695 P.2d 261.
Adequate Explanation
Also to preserve the judgment below, Chrysler proposes that the commission failed to sufficiently explain its TTD award as required by State ex rel. Mitchell v. Robbins & Myers, 6 Ohio St.3d at 483-484, 6 OBR at 534, 453 N.E.2d at 724 (commission must state evidence and give brief explanation for decision to grant or deny TTD). But rather than attack the commission’s cursory explanation, Chrysler uses the opportunity to resurrect its “no-evidence-of-causation” argument. We reject this argument for the reasons already discussed. We further conclude that while the instant commission order is less than comprehensive, it is at least as informative as the order in Mitchell, which was vacated due to lack of evidence, not for inadequate explanation. Id. at 484, 6 OBR at 534, 453 N.E.2d at 724. Accordingly, we cannot grant Chrysler any relief on this basis.
Reimbursement
Chrysler also insists that if we reverse in Garrett’s favor, it is still entitled to reimbursement from the Surplus Fund because the commission failed to appeal. Again, we disagree.
*170The source of Chrysler’s asserted right to reimbursement is former R.C. 4123.515,3 which provided, in part:
“[I]f the decision of the district hearing officer is appealed by the employer or the administrator, the bureau shall withhold compensation and benefits during the course of the appeal to the regional board of review, but where the regional board rules in favor of the claimant, compensation and benefits shall be paid by the bureau or by the self-insuring employer whether or not further appeal is taken. If the claim is subsequently denied, payments shall be charged to the surplus fund created under division (B) of section 4123.34 of the Revised Code, and if the employer is a state risk such amount shall not be charged to the employer’s experience and if the employer is a self-insurer such amount shall be paid to the self-insurer from said surplus fund.” (Emphasis added.) 137 Ohio Laws, Part II, 3934, 3939.
This statute allowed reimbursement only if compensation had been denied on appeal from the regional board of review. Accord State ex rel. Peabody Coal Co. v. Indus. Comm. (1989), 44 Ohio St.3d 104, 105, 541 N.E.2d 74, 75; State ex rel. Eaton v. Lancaster (1989), 44 Ohio St.3d 106, 541 N.E.2d 64; and Eaton, 40 Ohio St.3d at 416, 534 N.E.2d at 58 (Douglas, J., dissenting). Garrett was not denied compensation at subsequent levels of commission or judicial review; the TTD and medical benefit awards were left undisturbed. As former R.C. 4123.515 confers no right to reimbursement where the claimant prevails, Chrysler is not entitled to reimbursement under these facts.
Having found that some evidence exists for the commission’s decision and that none of Chrysler’s other arguments justifies the court of appeals’ judgment in Chrysler’s favor, we reverse the order granting a writ of mandamus and reinstate the commission’s award to Garrett of TTD and medical benefits.
Judgment reversed.
Douglas, F.E. Sweeney and Pfeifer, JJ., concur. *171Moyer, C.J., Cook and Lundberg Stratton, JJ., concur in part and dissent in part.. The commission’s orders do not specify reliance on the Kalb depositions, but the parties do not dispute this.
. R.C. 4123.515 was repealed effective October 20, 1993. On that same date, R.C. 4123.519 was amended and renumbered R.C. 4123.512, which provides that, “Any action pending in common pleas court or any other court ón January 1, 1986, under this section is governed by former sections * * * 4123.515 * * * and 4123.519 * * * of the Revised Code.” The present action was not pending in any court on January 1, 1986. Chrysler contends that any attempt to apply this legislation retroactively to preclude reimbursement would be unconstitutional. Because we find that former R.C. 4123.515 confers no right to reimbursement where the claimant ultimately prevails on the issue of entitlement to TTD, there is no need to address the retroactive application of R.C. 4123.512.