Opinion
Petitioner (applicant) seeks review of the decision of respondent Workers’ Compensation Appeals Board (Board) amending the findings of the workers’ compensation judge (WCJ) that applicant sustained industrial injuries to his cardiovascular system resulting in 100 percent permanent disability.
We hold the Board’s decision must be annulled because the Board erred in concluding the cardiovascular injuries were not industrial.
Factual Background and Procedural History
On March 13, 1986, applicant sustained a myocardial infarction (heart attack) while performing heávy work in the course of his employment as a truck driver for defendant Commercial Carriers, Inc. He sustained a second heart attack on March 31, 1986; and, while convalescing in Kaiser Hospital, he sustained cerebral vascular strokes on April 6 and 9, 1986. He is semicomatose and confined to a convalescent hospital.
Applicant did not testify because his semicomatose condition precludes speaking, reading and writing.
A CPK (creatine phosphokinase) heart-muscle enzyme test administered at the hospital revealed a CPK level of 2,200; and heparin (a blood thinner) was administered to applicant to prevent blood clots. An echocardiogram was also performed. Although the hospital physicians apparently concluded an embolus (small blood clot) from the heart caused the strokes, the hospital records do not specifically indicate that an embolus broke off the heart, traveled to the brain, and caused the strokes.
Dr. Gromis and Dr. Padova concluded the heart attacks were industrially related, but disagreed as to whether the strokes were related to the heart attacks. Dr. Gillis originally opined the heart attacks precipitated the strokes, but subsequently changed his opinion.
*250Dr. Gromis submitted several medical reports and testified at the trial. Dr. Gillis and Dr. Padova were deposed, but did not testify at the trial.
Dr. Gromis consistently opined it was reasonably probable that following the massive heart attacks an embolus traveled from applicant’s heart to the brain and caused strokes in both the right and left sides of the brain; although there is no medical test that can confirm this opinion, the opinion is based on experience, probabilities, and medical literature; the elevated CPK test level (2,200) of the heart muscle enzyme indicates the massiveness of the heart attack; the higher the enzyme level, the greater the likelihood of a stroke; and due to the brief time interval of several days between the second heart attack and the strokes, it is logical to conclude the offending embolus came from the heart.
Dr. Gillis, reporting for defendant, originally opined he could state no conclusion as to industrial causation of the heart attacks, but he concluded applicant “had unfortunate complications following his March, 1986, heart attack. Specifically, that heart attack progressed in degree later that month and then gave rise in April, 1986, to a stroke. Since that stroke [applicant] has been permanently and totally disabled . . . .”
Subsequently, Dr. Gillis again reported he was unable to determine whether the heart attacks were industrially related. Defendant then referred the matter to Dr. Padova.
Dr. Padova opined the heart attacks were industrially related, but the strokes were not related to the heart attacks, observing it is more likely the strokes were due to a thrombosis of the arteries. Dr. Padova concluded that although cerebral embolus secondary to intracardiac mural thrombus was suspected as a cause of the strokes, this diagnosis was not proven since an intracardiac thrombus was not identified in the echocardiography and since the usual complications associated with thrombosis formation, such as congestive heart failure, persistent hypotension, or prolonged intermittent cardiac arrhythmia, apparently did not occur after the heart attack.
Upon reviewing Dr. Padova’s report, Dr. Gillis retreated from his earlier opinion and concluded the strokes were not related to the heart attacks. In yet another report, Dr. Gillis stated that since the anticoagulation (heparin) treatment of the heart attacks at the hospital did not prevent the second stroke, there never was an embolic stroke, and hence there was no relation between the strokes and the heart attacks. He also noted the hospital echo-cardiogram showed no evidence of thrombus, and other cardiac manifestations, such as cardiac arrhythmia, can lead to embolic strokes.
*251At his deposition, Dr. Gillis agreed there was some objective evidence to support the existence of an embolism, and if one could show an embolus going from the heart, the stroke and its associated disability would be connected to the heart condition, but “there’s no proof positive. And I go on proof positive.” He stated further the hospital records indicate the treating physicians assumed there had been an embolus, but “if you use some Monday-morning quarterbacking and you look at the records and you say, ‘Well, let’s look at the criteria and were they met,’ you come to the conclusion that no one could ever prove there had been an embolic stroke.” Dr. Gillis discussed risk factors of thrombotic stroke such as age, high blood pressure history, smoking history, and cerebral arteriosclerosis, but omitted reference to heart attack as a risk factor in thrombotic stroke. He also discussed transient ischemic attacks as a harbinger of thrombotic stroke. Dr. Gillis stated he knew of no medical literature that says the worse the heart attack, the greater the likelihood of stroke.
In response to Dr. Gillis’s changed opinion, Dr. Gromis reported that regardless of Dr. Gillis’s speculation as to several mechanisms, such as cerebral arteriosclerosis, hypotension, and arrhythmia, that might explain the relationship between a massive heart attack and the stroke that follows it, the fact remains that applicant’s heart attack, which was as massive as can be observed (CPK heart muscle enzyme test level of 2,200), when considered in the context of applicant’s age and early history, caused the multiple strokes occurring soon after the heart attacks. In view of the 2,200 CPK level, an embolic etiology for the strokes is most likely, especially in light of the two well-documented strokes occurring on opposite sides of the brain soon after the heart attacks.
Dr. Gromis reported further there is a fundamental tenet in medicine that whenever possible, a single unifying diagnosis is more likely to be the correct diagnosis, whereas multiple unrelated diagnoses should be considered only as a last resort. It is common knowledge in medical literature that massive heart attacks go on to develop cerebral strokes soon thereafter.1 The diagnosis of embolic stroke is a clinical one and there is no medical test available, such as echocardiology, that can alter the diagnosis. The single unifying diagnosis here is that heart attacks lead to strokes, and when massive heart attacks lead to multiple strokes in opposite sides of the brain soon after the heart attacks, the unifying diagnosis is massive heart attacks *252with complications of embolic stroke. Dr. Gillis, on the contrary, disregards the single unifying diagnosis tenet in accounting for applicant’s heart attacks and complications, instead invoking three separate diagnoses: (1) Myocardial infarction, (2) right-sided thrombotic stroke, and (3) left-sided thrombotic stroke, all occurring within a period of three weeks. These separate diagnoses might be convincing if the time interval between the heart attack and the strokes was three years or more, but not where the interval is merely three weeks, as here.
Dr. Gromis reported further that Dr. Gillis engages in “backwards reasoning” and poor understanding of stroke disorders in concluding that since intravenous heparin did not prevent the second stroke applicant could not have had a cerebral embolus. Rather, it is likely that heparin therapy prevented additional multiple strokes, thereby saving applicant’s life.
As to Dr. Gillis’s statement he knows of no medical literature that says the worse the heart attack, the greater the likelihood of stroke, Dr. Gromis cited the article on myocardial infarction and stroke in the Neurology journal (see ante, fn. 1) wherein it is stated that stroke as a complication of acute myocardial infarction has been long recognized and patients with acute myocardial infarction are at increased risk of stroke during the first month of convalescence. Dr. Gromis also referred to statements in this article that 92 percent of patients with stroke following heart attacks had CPK levels greater than 1,160 (as compared to applicant’s CPK level of 2,200), and the incidence of stroke among patients with a CPK level of 1,160 is 24 times higher than among patients with lower CPK levels.
Regarding Dr. Gillis’s speculative discussion of cerebral arteriosclerosis, age, blood pressure history, and smoking history as possible risk factors for thrombotic strokes, Dr. Gromis noted Dr. Gillis omitted heart attack as a risk factor, if not in fact the most likely cause, of thrombotic stroke as noted in the cited article on myocardial infarction and stroke in the Neurology journal. As to Dr. Gillis mentioning transient ischemic attacks as a harbinger of a thrombotic stroke, Dr. Gromis noted there is no evidence applicant ever had a transient ischemic attack.
Dr. Gromis concluded his final report by stating that based upon his review of the medical literature, it is a medical certainty that applicant’s multiple strokes were temporally, directly and causally related to the antecedent massive acute myocardial infarctions; and, absent the myocardial infarctions, applicant’s disabling strokes would not have occurred.
Relying on Dr. Gromis’s opinion that the industrially caused heart attacks precipitated the strokes and on Dr. Gillis’s original opinion that the *253heart attacks precipitated the strokes, the WCJ found applicant sustained industrial injury to his cardiovascular system resulting in 100 percent permanent disability. The WCJ opined Dr. Gillis’s subsequent attempts to retreat from his original position as to causation of the strokes are not persuasive.
In his report on reconsideration, the WCJ again referred to Dr. Gillis’s original unequivocal opinion that the March 1986 heart attacks progressed in degree later that month and gave rise to the April 1986 stroke, and found persuasive Dr. Gromis’s rebuttal of Dr. Gillis’s deposition statements as to risk factors for thrombotic stroke, transient ischemic attacks, and heparin treatment. The WCJ reported that in view of the close proximity between the heart attacks and the strokes, and considering the tenuousness of Dr. Gillis’s explanation for his changed opinion as to the causal relationship between the heart attacks and strokes, the WCJ relied on the opinion of Dr. Gromis and the original opinion of Dr. Gillis in finding a causal connection between the heart attacks and the strokes; Dr. Gillis’s original opinion was more persuasive than his subsequent deposition rationalization for withdrawing his original opinion on the causation issue; Dr. Gillis’s modified conclusion is untenable; although the WCJ might agree with Dr. Gillis’s statement that no one could prove an embolic stroke by proof positive, the same statement would apply to a thrombotic stroke, one of which (embolic or thrombotic) the stroke must necessarily have been; and proof positive is not required in workers’ compensation cases, but rather the WCJ must ascertain which of the two medical theories is reasonably probable.
In a two-to-one decision, the Board granted reconsideration, concluded applicant failed to meet his burden of proving the strokes were industrially related, amended the WCJ’s findings, and found applicant sustained industrial myocardial infarctions, but the cardiovascular accidents (strokes) were not industrially related. The Board remanded the issue of permanent disability for redetermination at the trial level in light of the Board majority’s finding the strokes were not industrially related.
In its opinion and decision, the Board majority made a mere one-paragraph conclusional reference to Dr. Gromis’s earlier reports and a one-paragraph reference to Dr. Padova’s report, quoted a few paragraphs from Dr. Gillis’s final report and but two answers from his deposition testimony without any reference to Dr. Gillis’s two earlier reports, and made a mere three-sentence reference to Dr. Gromis’s final report and no reference to the medical literature.
Based on this very limited review of the record, the Board majority concluded: “We find Dr. Gillis’s opinion, as expressed in his deposition and *254in the quoted portions of his reports, to be more probative than Dr. Gromis’s. Dr. Gillis’s opinion establishes that the signs that would accompany a stroke that was attributable to the heart problems that applicant had, did not exist here. Thus, we agree with Dr. Gillis’s conclusion that Dr. Gromis’s opinion represents less than reasonable medical probability that the cardiovascular accidents were industrially related. Accordingly, we conclude that applicant has failed to meet his burden of establishing that the cardiovascular accidents were industrially related.”
One member of the three-member Board panel dissented without filing an opinion.
Discussion
1. Standard of appellate review.
In resolving the petition for writ of review, we must determine whether the evidence, when reviewed in the light of the entire record, supports the Board’s decision (Universal City Studios, Inc. v. Workers’ Comp. Appeals Bd. (1979) 99 Cal.App.3d 647, 656 [160 Cal.Rptr. 597]; see National Convenience Stores v. Workers’ Comp. Appeals Bd. (1981) 121 Cal.App.3d 420, 424 [175 Cal.Rptr. 378]); and in doing so, we must consider the weight or persuasiveness of all the evidence, not just whether there is substantial evidence in favor of the respondent. (Skip Fordyce, Inc. v. Workers’ Comp. Appeals Bd. (1983) 149 Cal.App.3d 915, 920 [197 Cal.Rptr. 626].)
We are not bound to accept the Board’s factual findings where they are illogical, unreasonable, or improbable (Insurance Co. of North America v. Workers’Comp. Appeals Bd. (1981) 122 Cal.App.3d 905, 911 [176 Cal.Rptr. 365]), where they do not withstand scrutiny when considered in light of the entire record (Duke v. Workers’ Comp. Appeals Bd. (1988) 204 Cal.App.3d 455, 460 [251 Cal.Rptr. 185]), or where on a case-by-case examination we discern an inequitable result when the record is examined for fairness, reasonableness and proportionality in the overall scheme of the workers’ compensation law and the purposes sought to be accomplished by that law. (National Convenience Stores v. Workers’ Comp. Appeals Bd., supra, 121 Cal.App.3d at p. 424; Universal City Studios, Inc. v. Workers’ Comp. Appeals Bd., supra, 99 Cal.App.3d at pp. 658-659.)
Finally, where the Board’s decision is not within the realm of what a reasonable trier of fact could find, the decision is not supported by substantial evidence and must be annulled. (Skip Fordyce, Inc. v. Workers’ Comp. *255Appeals Bd., supra, 149 Cal.App.3d at p. 921; Insurance Co. of North America v. Workers’ Comp. Appeals Bd., supra, 122 Cal.App.3d at p. 911.)
2. Standard of review by Board on reconsideration.
Unlike the court, the Board is empowered on reconsideration to resolve conflicts in the evidence, to make its own credibility determinations, and to reject the findings of the WCJ and enter its own findings on the basis of its review of the record; nevertheless, any award, order or decision of the Board must be supported by substantial evidence in the light of the entire record. (Lamb v. Workmen’s Comp. Appeals Bd. (1974) 11 Cal.3d 274, 280-281 [113 Cal.Rptr. 162, 520 P.2d 978]; Garza v. Workmen’s Comp. App. Bd. (1970) 3 Cal.3d 312, 317 [90 Cal.Rptr. 355, 475 P.2d 451]; LeVesque v. Workmen’s Comp. App. Bd. (1970) 1 Cal.3d 627 [83 Cal.Rptr. 208, 463 P.2d 432].)
The foregoing standard is not met by simply isolating evidence which supports the Board and ignoring other relevant facts of record which rebut or explain the evidence. (Lamb v. Workmen’s Comp. Appeals Bd., supra, 11 Cal.3d at p. 281; see Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at p. 317; City of Santa Ana v. Workers’ Comp. Appeals Bd. (1982) 128 Cal.App.3d 212, 219 [180 Cal.Rptr. 125]; Western Electric Co. v. Workers’ Comp. Appeals Bd. (1979) 99 Cal.App.3d 629, 645 [160 Cal.Rptr. 436].)
Thus, in relying on the opinion of a particular physician in making its determination, the Board may not isolate a fragmentary portion of the physician’s report or testimony and disregard other portions that contradict or nullify the portion relied on; the Board must give fair consideration to all of that physician’s findings. (City of Santa Ana v. Workers’ Comp. Appeals Bd., supra, 128 Cal.App.3d at p. 219.) As stated in Gay v. Workers’ Comp. Appeals Bd. (1979) 96 Cal.App.3d 555, 564 [158 Cal.Rptr. 137], in evaluating the evidentiary value of medical evidence, a physician’s report and testimony must be considered as a whole rather than in segregated parts; and, when so considered, the entire report and testimony must demonstrate the physician’s opinion is based upon reasonable medical probability. (See Lamb v. Workmen’s Comp. Appeals Bd., supra, 11 Cal.3d at p. 281; McAllister v. Workmen’s Comp. App. Bd. (1968) 69 Cal.2d 408, 416-417 [71 Cal.Rptr. 697, 445 P.2d 313].) Hence, the Board may not blindly accept a medical opinion that lacks a solid underlying basis and must carefully judge its weight and credibility. (National Convenience Stores v. Workers’ Comp. Appeals Bd., supra, 121 Cal.App.3d at p. 426.)
“ ‘[N]ot all medical opinion constitutes substantial evidence upon which the Board may rest its decision. Medical reports and opinions are not *256substantial evidence if they are known to be erroneous, or if they are based on . . . incorrect legal theories . . . [or] on surmise, speculation, conjecture or guess.’ ” (Insurance Co. of North America v. Workers’ Comp. Appeals Bd., supra, 122 Cal.App.3d at p. 917; see Hegglin v. Workmen’s Comp. App. Bd. (1971) 4 Cal.3d 162, 169 [93 Cal.Rptr. 15, 480 P.2d 967]; Place v. Workmen’s Comp. Appeals Bd. (1970) 3 Cal.3d 372, 378 [90 Cal.Rptr. 424, 475 P.2d 656]; Guerra v. Workers’ Comp. Appeals Bd. (1985) 168 Cal.App.3d 195, 199 [214 Cal.Rptr. 58]; Baptist v. Workers’ Comp. Appeals Bd. (1982) 137 Cal.App.3d 903, 906 [187 Cal.Rptr. 270].)
When the WCJ’s finding of compensable injury is supported by solid, credible evidence, it is to be accorded great weight by the Board and should be rejected only on the basis of contrary evidence of considerable substantiality. (Lamb v. Workmen’s Comp. Appeals Bd., supra, 11 Cal.3d at p. 281; Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at pp. 318-319; Los Angeles Unified School Dist. v. Workers’ Comp. Appeals Bd. (1981) 116 Cal.App.3d 393, 400 [171 Cal.Rptr. 841]; Western Electric Co. v. Workers’ Comp. Appeals Bd., supra, 99 Cal.App.3d at p. 642.) The WCJ’s findings on credibility are entitled to great weight because the WCJ has the opportunity to observe the demeanor of the witnesses and weigh their statements in connection with their manner on the stand. (Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at pp. 318-319; Western Electric Co. v. Workers’ Comp. Appeals Bd., supra, 99 Cal.App.3d at p. 643.)
When the WCJ has relied upon solid, credible evidence to make a wholly supported finding of compensability, the Board may not avoid its obligation to give great weight to that finding by ignoring the evidence on which it is based, since, in doing so, the Board renders a decision which is not supported by substantial evidence in the light of the entire record. (Lamb v. Workmen’s Comp. Appeals Bd., supra, 11 Cal.3d at p. 283.)
Factual findings of the Board are not supported by substantial evidence in light of the entire record where such findings are based upon inferences which cannot be fairly drawn from the evidence, based on evidence lacking probative force, or based on a purely “ ‘fanciful conclusion.’ ” (Insurance Co. of North America v. Workers’ Comp. Appeals Bd., supra, 122 Cal.App.3d at pp. 910-911.)
3. Board majority erred in amending findings.
Applying the cited authorities on the Board’s power to reject the WCJ’s findings and substitute its own findings, we first note the Board majority patently disregarded the mandate that it review the evidence in the light of the entire record. (Lamb v. Workmen’s Comp. Appeals Bd., supra, *25711 Cal.3d at pp.280-281; Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at p. 317; LeVesque v. Workmen’s Comp. App. Bd., supra, 1 Cal.3d at p. 637.) Instead, the Board majority erroneously relied on isolated fragmentary portions of Dr. Gillis’s deposition and later report while disregarding substantial evidence that tended to contradict the isolated fragments relied on, such as Dr. Gillis’s original report, almost all of Dr. Gromis’s extensive reports and testimony expressing his opinion as a reasonable probability, and the medical literature showing the causal relationship between acute heart attacks and strokes occurring soon thereafter. (City of Santa Ana v. Workers’ Comp. Appeals Bd., supra, 128 Cal.App.3d at p. 219; see Lamb v. Workmen’s Comp. Appeals Bd., supra, 11 Cal.3d at p. 281; Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at p. 317; Western Electric Co. v. Workers’ Comp. Appeals Bd., supra, 99 Cal.App.3d at p. 645.)
Moreover, Dr. Gillis’s changed opinion in his deposition and final report was to a large extent based on surmise, speculation, conjecture, and guess, and thus did not constitute substantial evidence upon which the Board majority could rest its decision. (Insurance Co. of North America v. Workers’ Comp. Appeals Bd., supra, 122 Cal.App.3d at p. 917; see Hegglin v. Workmen’s Comp. App. Bd., supra, 4 Cal.3d at p. 169; Place v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at p. 378; Guerra v. Workers’ Comp. Appeals Bd., supra, 168 Cal.App.3d at p. 199; Baptist v. Workers’ Comp. Appeals Bd., supra, 137 Cal.App.3d at p. 906). The Board majority could not blindly accept the changed opinion of Dr. Gillis, which lacked a solid, underlying basis; it should have weighed Dr. Gillis’s changed opinion and his credibility against the contrary medical evidence relied upon by the WCJ and the credibility of Dr. Gromis, who testified at the trial and was found credible by the WCJ. (National Convenience Stores v. Workers’ Comp. Appeals Bd., supra, 121 Cal.App.3d at p. 426; see Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at p. 319.) The evidence relied upon by the Board simply did not constitute contrary evidence of considerable substantiality to warrant the Board’s rejection of the WCJ’s findings which were based on the solid, credible evidence evinced in Dr. Gromis’s reports, Dr. Gillis’s original report, and the medical literature. (Lamb v. Workmen’s Comp. Appeals Bd., supra, 11 Cal.3d at p. 283; see Garza v. Workmen’s Comp. App. Bd., supra, 3 Cal.3d at p. 319.)
We are not bound to accept the Board’s factual findings where, as here, they are unreasonable (Insurance Co. of North America v. Workers’ Comp. Appeals Bd., supra, 122 Cal.App.3d at p. 911), are not within the realm of what a reasonable trier of fact could find (ibid.; Skip Fordyce, Inc. v. Workers’ Comp. Appeals Bd., supra, 149 Cal.App.3d at p. 921), do not withstand scrutiny when viewed in the light of the entire record (Duke v. Workers’ Comp. Appeals Bd., supra, 204 Cal.App.3d at p. 460), and lead to an *258inequitable result when the entire record is examined for fairness, reasonableness, and proportionality in the overall scheme of the workers’ compensation law and the purposes sought to be accomplished by that law. (National Convenience Stores v. Workers’ Comp. Appeals Bd., supra, 121 Cal.App.3d at p. 424; Universal City Studios, Inc. v. Workers’ Comp. Appeals Bd., supra, 99 Cal.App.3d at pp. 658-695.)
For the foregoing reasons, we conclude the Board erred in amending the WCJ’s findings.
Disposition
The opinion and decision of respondent Board after reconsideration is annulled; and the matter is remanded for further proceedings consistent with the views expressed herein.
Danielson, J., and Croskey, J., concurred.