The question submitted for resolution is whether the Workers’ Compensation Commission’s finding that appellant’s cerebral vascular condition was not related to her compensable injury of February 20, 1976, is supported by substantial evidence.
It is undisputed that on February 20, 1976, appellant, while in the course of her employment, sustained an injury to her left knee while attempting to move a desk in the surgery department of the Conway Memorial Hospital. Dr. John Smith, a Conway physician, was immediately summoned to render medical assistance to the claimant. Dr. Smith prescribed Talwin as a pain medication. Three days later, while still in pain and on crutches, claimant consulted her family physician, Dr. Bob Bannister. Dr. Bannister immediately referred claimant to Dr. Leighton Millard, a Little Rock orthopedic surgeon. Dr. Millard concluded that claimant had sustained a torn cartilage in her knee joint and recommended surgery.
A few days after surgery, appellant experienced double vision, difficulty in focusing on objects, slurred speech, in-coordination, her tongue and throat felt paralized, could not sit up, nor hold her head up, could not walk, could not control her facial expressions, realized difficulty in lifting her arms, developed shortness of breach and could not feed herself. Emotional problems encountered by the claimant may be briefly characterized as crying and laughing hysterically.
Dr. Millard concluded that claimant had suffered a cerebral vascular accident and referred claimant to Dr. Robert Galbraith, a Little Rock neurologist.
The Commission, in a split decision, found that the claimant failed to prove by a preponderance of the evidence that the cerebral vascular accident was related to her compensable injury. This finding is predicated upon the opinion of Dr. Thomas M. Fletcher, a Little Rock neurosurgeon, who examined claimant, at the request of appellee-respondent, *682approximately one year after claimant sustained her injury. Dr. Fletcher’s opinion may be briefly summarized by quoting from his medical report of February 28, 1977:
“After examining this patient and reviewing the history notes and examination of the hospitalization at Baptist Hospital from February 29,1976 to March 16,1976, it is my opinion that the neurological impairment that she sustained very likely was due to cerebral embolism possibly related to her thrombophlebitis, although this has not been definitely confirmed, it would appear to be the most likely probability.” (Emphasis added)
We believe an analysis of Dr. Fletcher’s opinion within the context of relevant factual matters demonstrates the hollowness of Dr. Fletcher’s theory of the cause of claimant’s cerebral vascular accident. We are persuaded that the dissenting opinion of one of the Commissioners illustrates the insubstantiality of Dr. Fletcher’s report and, accordingly, we quote from a portion of this opinion:
“The one overriding factor that the majority ignores in this case is the exact time relationship between the surgery required by the injury and the numerous symptons experienced by the claimant immediately after the surgery.
Despite the fact that the medical evidence of record evinces equal support for the two most probable causative factors (i.e., (1) the injury and/or surgery to correct the injury and (2) a condition of thrombophlebitis, the only evidence of which was of a questionable nature) the majority chose to select the factor separated by a time element of one year, rather than the factor having an immediate time relationship.
“Claimant had never, before the surgery, experienced such things as double vision, dizziness, speech difficulties, loss of muscle tone, shortness of breath, and, as described by Dr. Galbraith, a ‘labile emotional problem’ that did ‘reflect frontal lobe injury.’ Yet, immediately after regaining consciousness from surgery, *683claimant began to experience these symptoms. Only Dr. Fletcher places more emphasis on the supported thrombophlebitis condition of one year earlier as the more likely cause of claimant’s condition after surgery, and the majority’s decision obviously relied upon Dr. Fletcher’s opinion as a basis for denying this claim. Yet, the existence of a thrombophlebitic condition one year before claimant’s ‘cerebral vascular accident’ was never conclusively established, thereby creating even more question as to the value of Dr. Fletcher’s conclusion. This is indicated by Dr. Fletcher’s deposition at page 12 where the following testimony appears with reference to a report which Dr. Fletcher had read from claimant’s personal physician of one year earlier:
‘Well, I think the fact that he placed her on anticoagulants would be a strong implication that he thought she did have a thrombophlebitis in October . . .’ (Emphasis added)
“It should suffice to say, even though additional diagnostic testing might have assisted in arriving at a more definite conclusion, that the physicians were unanimous, in their feeling that the claimant’s post-surgery symptoms and clinical findings were consistent with a classical type of cerebral vascular accident.
“Further, and of even greater significance than Dr. Fletcher’s supposition concerning pre-existing thrombophlebitis, is the fact that Dr. Millard, who performed claimant’s surgery, after which the above described symptoms arrived, stated that there was ‘mo evidence of thrombophlebitis’ (emphasis added) during the period of his treatment of the claimant.”
Dr. Millard testified on direct examination:
Q. On April 22,1976, Doctor, I believe you wrote a letter to Mr. Hill, (insurance adjuster) and on that date you expressed an opinion as to the cause of her entire problem.
*684A. Yes, sir, I did. I felt at that time and still feel that the problem was related to the straining injury that she had at work.
Dr. Fletcher testified on cross-examination:
Q. Doctor, is it your opinion that the injury of Mrs. Watson sustained on February 20, 1976, could have precipitated or have a triggering effect on her thrombophlebitis condition?
A. It’s possible that it could—that it could aggravate it or make it apparent or by the stress of the stretching out the leg, that it could cause a sudden pain.
Q. Do you have an opinion as to whether or not, or as to whether the cerebral embolism occurred subsequent to the February 20, 1976, injury?
A. The February 20th was the date that she hurt her knee.
Q. Yes, sir. She was working at the time.
A. Well, I do not believe that—I mean, I would think that it would have occurred after that time, sometime after that time.
Q. And the resulting symptoms which you’ve described would also, of course, have occurred after February 20, 1976?
A. Yes, because Dr. Bannister doesn’t mention them in his letter as being noticed at that time.
It is settled law that a decision of the Workers’ Compensation Commission will be affirmed if there is substantial evidence to support it. On the other hand, if the evidence falls short of substantiality, appellate courts are duty bound to reverse. What constitutes substantial evidence is a question of *685law. Boyd Excelsior Fuel Co. v. McKown, 226 Ark. 174, 288 S.W. 2d 614 (1956); Cummings v. United Motor Exchange, 236 Ark. 735, 368 S.W. 2d 82 (1963). After a careful examination of the record and applicable law, we are persuaded that the Commission’s holding is not supported by substantial evidence.
It is universally recognized that our Workers’ Compensation Law should be liberally construed, as opposed to a strict interpretation, and that doubtful cases and all disputed evidence should be resolved in favor of the claimant. Boyd Excelsior Fuel Co. v. McKown, supra; Eddington v. City Electric Co., 237 Ark. 804, 376 S.W. 2d 550 (1964); Aluminum Co. of America v. Henning, 260 Ark. 699, 543 S.W.2d 480 (1976); Jacob Hart Seed Co. v. Thomas, 253 Ark. 176, 485 S.W. 2d 200 (1972); Southwestern Bell Telephone Co. v. Brown, 256 Ark. 54, 505 S.W. 2d 207 (1974). If this ruleof law has any meaning, force or effect, it would, indeed, unduly tax the imagination to conceive of a circumstance requiring its application more so than in the instant case.
In American Can Co., Employer v. Charles McConnell, Employee, 266 Ark. 741, 587 S.W. 2d 583 (Ark. App. 1979), the Arkansas Court of Appeals made the following relevant observation regarding the necessity of conclusive medical evidence:
‘To appraise the true degree of indispensability which should be accorded medical testimony, it is first necessary to dispel the misconception that valid awards can stand only if accompanied by a definite medical diagnosis. True, in many instances, it may be impossible to form a judgment on the relation of the employment to the injury, or relation of the injury to the disability, without analyzing in medical terms what the injury or disease is. But this is not invariably so. In appropriate circumstances, awards may be made when medical evidence on these matters is inconclusive, indecisive, fragmentary or even nonexistent.’’ (Emphasis supplied)
While we are convinced that there is ample, competent medical evidence to establish a causal relationship between appellant’s knee injury and her cerebral vascular accident, it *686is plain that appellant sustained a compensable injury necessitating surgery to her left knee; and that she had never, prior to the surgery, experienced the numerous problems and difficulties detailed in this record.
Accordingly, the decision of the Arkansas Workers’ Compensation Commission is reversed and this cause is remanded to the Commission for further proceedings consistent with this opinion.
Newbern and Penix, JJ., dissent.