Setting the stage for a scientific and regulatory battle, Merck said yesterday that a potential successor to its painkiller Vioxx had shown in a large study to be no more dangerous to the heart than an older painkiller still widely in use.

The Merck drug, Arcoxia, is already sold in 62 countries, but so far regulators in the United States have deferred approving the drug. Until yesterday analysts had said that Arcoxia had little chance of being approved because of concerns that it might have heart risks similar to Vioxx, which Merck withdrew from the market in 2004 over safety concerns.

But in announcing the study’s main results yesterday, Merck indicated that it intended to push for Food and Drug Administration approval to sell Arcoxia in this country.

Some independent scientists, though, criticized the company for making its announcement without releasing the study’s full details. And they noted that the Arcoxia side effects that Merck did disclose yesterday included high blood pressure.

At stake is a potentially huge market. Tens of millions of Americans regularly use anti-inflammatory painkillers for arthritis.

Merck sold almost $3 billion of Vioxx a year before it withdrew the drug in September 2004 after a clinical trial showed that Vioxx sharply increased patients’ risks of heart attacks and strokes. Merck now faces more than 14,000 lawsuits from people who say that they suffered heart damage after taking Vioxx.

And yet a drug in the same class as Vioxx — Celebrex, made by Pfizer — continues to have annual sales of nearly $2 billion, despite carrying an F.D.A.-ordered warning label that it, too, may pose heart risks.

If Arcoxia is approved, Merck could regain some of the profits it lost when it withdrew Vioxx — money that could conceivably help the company pay the legal bills it faces over Vioxx.

The new study, which covered 34,000 patients, shows that Arcoxia’s risks are comparable to the older painkiller, diclofenac, Merck officials said.

Diclofenac, a prescription drug sold under brand names that include Cataflam and Voltaren, is not popular in this country, in part over concerns that it causes liver problems. But it is widely used in Europe.

Merck is pursuing a high-risk strategy in pressing ahead with its application to sell Arcoxia in the United States, scientists said. Accusations that Merck did not tell doctors everything it knew about the risks of Vioxx have damaged the company’s reputation. Merck must now persuade skeptical regulators and doctors that it is fully disclosing Arcoxia’s potential hazards.

Dr. Steven E. Nissen, the president of the American College of Cardiology, said Merck should have disclosed more information from the Arcoxia trial or not discussed it at all.

“The bottom line here — they didn’t really release enough data for us to make a conclusion,” said Dr. Nissen, who is overseeing a large clinical trial sponsored by Pfizer to study the heart risks of Celebrex. “It was a very carefully crafted press release.”

Dr. Bruce Psaty, a professor of medicine and epidemiology at the University of Washington, said that the trial would have been more meaningful if Merck had used naproxen — sold over the counter as Aleve — as a comparison drug, rather than diclofenac.

Dr. Sean Curtis, who heads the Arcoxia development program for Merck, said the company would disclose the study’s full results before the end of 2006, both in a peer-reviewed scientific journal and at a medical conference.

“We really want to present the entirety of the data after peer review,” Dr. Curtis said.

Diclofenac is an appropriate comparison drug for the study because it is so widely used worldwide, he said.

The Merck study, called Medal, included results from three clinical trials comparing Arcoxia and diclofenac. Patients took the drugs for an average of 17 months, and 10,000 patients took them for two years or more, Merck said.

On average, patients who took Arcoxia were slightly less likely to have a heart attack or other serious cardiovascular problem than those who took diclofenac, according to the company.

But patients on Arcoxia were significantly more likely to have hypertension or swelling, Merck said. Hypertension is a particularly troubling side effect, doctors said, because it is associated with heart attacks and congestive heart failure.

Arcoxia and Vioxx, as well as Celebrex, are part of a class of drugs developed in the 1990’s called cox-2 inhibitors. Older painkillers like ibuprofen inhibit the production of two enzymes, cox-1 and cox-2, that cause inflammation and pain. But cox-1 enzymes also protect the stomach and taking the older medicines for long periods of time can lead to stomach pain or gastrointestinal bleeding.

Drugs like Vioxx and Celebrex were specifically formulated to suppress the cox-2 enzyme without affecting cox-1. Scientists hoped that they would provide pain relief without damaging the stomach.

But clinical trials have shown that Celebrex and Vioxx appear to cause many more heart attacks and strokes than the older medicines. The reason is still open to debate. But many cardiologists now argue that cox-2 is crucial to the proper functioning of blood vessels and that inhibiting it may increase the risk of dangerous blood clots.

Another question that remains unanswered is whether the cox-2 inhibitors actually prevent stomach damage. In clinical trials, only Vioxx has been shown to reduce stomach bleeding compared with older painkillers.

In reporting the main findings of the Medal study yesterday, Merck said that patients on Arcoxia had significantly fewer stomach and liver problems than those on diclofenac. But it did not explain whether the difference resulted from an excess of liver problems, stomach problems or both.

Dr. Sidney Wolfe, the director of health research for the consumer advocacy group Public Citizen, said Merck needed to provide much more information about the results of the trial.

“They’re not presenting data or anything, they’re giving you a little hint,” Dr. Wolfe said. “There are more unanswered questions here than answered questions.”

Dr. Psaty, the University of Washington professor, said that because diclofenac suppressed the cox-2 enzyme more strongly than other older painkillers, it could have some of the same heart risks as Arcoxia and other new cox-2 drugs.

Unless a government agency carries out a large independent trial, patients and doctors may never know which painkillers are the safest, Dr. Psaty said.