Opinion
No. 9405671
August 17, 1998
I. INTRODUCTION
This case comes before the court on defendants' Motion in Limine to Exclude Certain Expert Testimony. In their motion defendants seek to exclude the trial testimony of John G. Keller, Ph.D., and Leon I. Charash, M.D., regarding the long term effects of a Prozac overdose on plaintiff Shea Hammond, as unreliable under the holdings of Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993), and Commonwealth v. Lanigan, 419 Mass. 15 (1994). The motion was argued on March 20, 1998; the parties submitted written materials but presented no testimony. For the following reasons defendants' motion is ALLOWED without prejudice.
II. FACTS
From the parties' submissions, the court summarizes the facts.
A. Shea Hammond's Medical Condition
Plaintiff Shea Hammond (Shea) was born November 3, 1989. Since that time, he has been diagnosed with a variety of medical conditions. At birth, Shea was diagnosed with persistent fetal circulation which resolved within a week. At nine months, he was diagnosed with cystinosis, a rare genetic disorder whereby the body fails to metabolize cystine. This failure results in the build up of cystine crystals in the body's organs which in turn causes irreversible damage. At age 13 months, Shea began phospho-cysteamine therapy to reduce his cystine levels.
Shea also suffered from chronic vomiting, which resulted in a gastronomy tube (G-Tube) being placed into his stomach for direct feeding. In February 1992, Shea was diagnosed with gastro-intestinal reflux and antroduodenal dysmotility, the failure of the intestines to push food through the system. As a result, at twenty-eight months a jejunostomy tube (J-tube) was placed into Shea's small intestine to bypass the stomach for feeding. As aresult of the J-tube insertion, Shea's vomiting substantially abated until October 1992 when he suffered a Yersinia bacterial infection. Shea was admitted to Newton-Wellesley Hospital for the infection and then transferred to Massachusetts General Hospital(MGH). At MGH a central intravenous line was surgically placed into Shea's chest to ensure that he could receive nutrition in an alternative method to the J or G tubes. In December 1992, while hospitalized at MGH, Shea was diagnosed with the brain condition Arnold-Chiari Malformation Type I. On January 13, 1993, Shea underwent brain surgery to correct this condition.
Arnold-Chiari Malformation Type I is a brain condition whereby the cerebellum is herniated so that portions of the brain protrude through an opening at the base of the skull.
On February 16, 1993, then three years and three months old, Shea was discharged form his long stay at MGH. At that time, the amount and strength of formula he could take through the J-tube was drastically decreased form what it had been the previous fall, so that he received the majority of his nutrition through his central line. On the way home for MGH, the Hammonds went to CVS to fill Shea's prescriptions. One of the prescriptions, for cimetidine, commonly known as Tagamet, was improperly filled by defendant Colleen Karjane with the anti-depressant Prozac.
Prozac is also known as fluoxetine and the court uses these terms interchangeably.
Unaware of the mistake, the Hammonds administered to Shea three doses of Prozac over the course of approximately eighteen hours, each approximately 48 Mg. After each administration, Shea vomited, became lethargic and exhibited neurological symptoms. In addition, there was one noted instance of intestinal bleeding. On February 17, 1993, the pharmaceutical error was discovered and the Hammonds took Shea to the Emergency Room at MGH where he was admitted for overnight observation. Shea was discharged the next day, February 18, 1993.
A typical adult dose of Prozac is 20-60 mg. See Gerald K. McEvoy, Pharm. D. Editor, AHFS Drug Information 97, p. 1705 [hereinafter AHFS].
On February 24, 1993, Shea visited his treating physician, Julie Ilfinger, M.D., at which time she noted that he appeared to be doing well. In addition, Shea increased the strength of the formula that he was receiving through the J tube from one-half to three-quarters strength. On March 8, 1993, Shea was again admitted to the hospital, this time for both neurological symptoms including right-side weakness, ataxia and abnormal gait, and GI symptoms including severe vomiting. Shea's GI symptoms did not improve until sometime in mid-1993. Until that time, Shea was nourished solely through his central line.
The limitation to feeding through the central line affected the treatment of Shea's cystinosis because, at that time, phospho-cysteamine could not be administered through the central line. Shea was unable to continue this therapy for a period of months. As a result, an increase in cystine crystals was discovered in his cornea, duodenum, rectum and bone marrow. Among other things, plaintiffs believe that this increase in cystine crystals has increased Shea's risk of renal failure.
B. Expert Opinion I. John G. Keller, Ph.D.
John G. Keller, Ph.D. is a toxicologist with extensive experience in government, industry and research projects in toxicology. In preparation for giving his opinion in this case, Dr. Keller reviewed: case records and depositions; medical records of Shea Hammond's stays at Newton-Wellesley Hospital, MGH and the National Institutes of Health; and medical literature studying the effects of Prozac on the human body and the effects of intravenous cysteamine therapy on patients with cystinosis.
Dr. Keller is expected to testify that the overdose of Prozac had both short and long term effects on Shea. Dr. Keller is of the opinion that Shea exhibited both acute symptoms of GI and neurological distress immediately after the ingestion of Prozac and longer term GI injuries that exacerbated his preexisting GI dismotility and prohibited uninterrupted treatment with cysteamine. Specifically, Dr. Keller suggests that the ingestion of Prozac affected the physiology of Shea's intestines and caused his long term symptoms.
ii. Leon I. Charash. M.D.
Leon I. Charash, M.D. is a pediatric neurologist. It is Dr. Charash's opinion that Shea's Prozac overdose exacerbated Shea's preexisting GI problems. For his opinion, Dr. Charash reliesupon: a review of Shea's medical records; the high dosage ofProzac ingested; the "known" adverse side effects of Prozac including dysphagia, esophagitis, gastritis, glossitis, and stomatitis; Shea's reported symptoms after ingestion of the drug; and the temporal relationship between Shea's symptoms and the overdose.
C. Medical Literature
Plaintiffs offered fourteen articles or excerpts from medical literature that they contend support the opinions of their experts. Thirteen of the articles discuss the effects and side effects of treatment with and/or overdose of Prozac. The remaining article, a case report on Shea Hammond, demonstrates the effectiveness of intravenous cysteamine therapy in reducing leukocyte cystine levels. The literature submitted does not discuss the diagnoses or typical course of Cystinosis, gastrointestinal reflux, antroduodenal dysmotility or Arnold-Chiari Malformation Type I, all conditions from which Shea suffered.
i. Literature on the Effects of Prozac
In contrast to Shea Hammond, the subject patients of the articles regarding Prozac generally were in good physical health, although many were being treated for psychological disorders. "[C]linical experience with fluoxetine HCL in patients with concomitant diseases or conditions is limited." Thus, Shea's particular medical problems limit direct application of the medical literature to this case.
Mosby's Complete Drug Reference 1997 Physicians GenRx-Drug Information II-934, [hereinafter Mosby's Drug Reference].
Prozac is a drug generally prescribed for treatment of both psychological disorders, such as depression, and behavioral conditions, such as obsessive-compulsive disorder and Tourette's Syndrome. Typical dosages for adolescents in trials studies range from 10 to 40 mg/d. In addition, the half-life or elimination rate of fluoxetine and its principal metabolite norfluoxetine has been documented to be relatively long, with multiple administrations of the drug reportedly prolonging the clearance rate. Indeed, because of the drug's long half-life it is recommended that contraindicated drugs not be administered until five weeks after the discontinuance of fluoxetine therapy.
John Piancentini, Ph.D., et.al., Psychopharmacologic Treatment of Child and Adolescent Obsessive Compulsive Disorder, Pediatric Psychopharmacology, 1992; 15:87-102.
AHFS at 1697.
Mosby's Drug Reference at II-933. See also AHFS at 1703.
Side effects from treatment with Prozac are generally mild and transient, most frequent of which are headache, nausea, vomiting, insomnia and tremor. Of the GI effects, nausea is the most common and usually subsides after a few weeks of use.
C. Lindsay DeVane, Pharm.D. and Floyd R. Sallee, M.D., Ph.D., Serotonin Selective Reuptake Inhibitors in Child and Adolescent Psychopharmacology: a Review of Published Experience, J. Clinical Psychiatry 57:2, February 1996, pp. 56 and 62-63.
Id.
Other adverse GI effects associated with fluoxetine therapy include abdominal pain and change in taste perception, which occur in approximately 3 and 2% of patients respectively . . . Vomiting, melena and flatulence reportedly occur in about 2% and gastroenteritis in about 1% of patients receiving the drug . . . [A]phthous stomatitis, dysphagia, eructation, esophagitis, gastritis, gingivitis, glossitis, melena, stomatitis, and thirst, have been reported in less than 1% of fluoxetine-treated patients; however a causal relationship has not been established. Bloody diarrhea, GI hemorrhage, colitis, duodenal or gastric ulcer, enteritis, pancreatitis, [and] fecal incontinence . . . have occurred rarely, but have not been definitely attributed to fluoxetine. (Emphasis supplied).
AHFS at 1700. See also Mosby's Drug Reference at II-935, (listing the same GI symptoms as reported by patients taking fluoxetine but coming to no conclusion as to cause and effect).
The most common nervous system effects of fluoxetine therapy include anxiety, nervousness, insomnia, drowsiness, fatigue or asthenia, tremor and dizziness.
AHFS at 1699. See also Mosby's Drug Reference at II-935.
Prozac overdose has been associated with agitation. severe vomiting, hypomania, seizures, and other signs of central nervous system excitation. At least two deaths have been attributed to fluoxetine overdose when the drug has been ingested in combination with other drugs, and at least one deal has been attributed to a massive overdose of fluoxetine alone. As with regular treatment with fluoxetine, reported symptoms from overdose "when there were no coingestants. were mild and transient . . . [In one study,] 48.6% of all patients that ingested fluoxetine alone remained asymptomatic[,]" and in documented cases of pediatric overdose, "[e]ighteen out of twenty patients (90%) remained asymptomatic after a minimum of 24 hours." The mean dosage taken by these pediatric patients was 23.4 mg, an amount substantially lower than Shea's reported ingestion of 144 mg.
Timothy P. Rohrig, et al., Fluoxetine Overdose: A Case Report, J. Anal. Toxicol., 1989; 13:305 [hereinafter Rohrig, Fluoxetine Overdose]; and Mosby's Drug Reference at II-936.
Rohrig, Fluoxetine Overdose: A Case Report, 13:305.
Raymond H. Feierabend, Jr., M.D., Benign Course in a Child with a Massive Fluoxetine Overdose, J. FAM. PRACT. 1995; 41:290.
Douglas J. Borys, BS, et al., Acute Fluoxetine Overdose: A report of 234 Cases, Am. J. Em. Med. 1992; 10:115.
Id.
In a reported case of a massive fluoxetine overdose in a four-year-old girl, however, no permanent effects were observed. Upon admission to the emergency room, the child exhibited hypervigilance, moderate psychomotor agitation and regular tachycardia. No GI symptoms were noted and vomiting was induced. Close observation of the child was maintained for approximately 72 hours during which time all psychomotor symptoms resolved. The child's parent reported that she "gradually returned to her normal state after about 2 weeks of appearing 'hyped up.'" At hospital admission, the fluoxetine serum level of the child in this case was reported at 3080 ng/mL, over ten times Shea's reported serum level of 235 ng/mL.
Feierabend, Benign Course in a Child with a Massive Fluoxetine Overdose at 41:289-291.
Id. at 290.
ii. Literature on Intravenous Cysteamine Therapy
In addition to their submissions regarding Prozac, plaintiffs submitted a published case report by the National Institutes of Health conducted on Shea Hammond documenting successful administration of cysteamine therapy through Shea's intravenous line. The case study notes that Shea was suffering from documented GI dysmotility and frequent vomiting which was unsuccessfully treated. The study does not attribute his dysmotility
William A. Gahl, et al., Intravenous Cysteamine Therapy for Nephrovathic Cystinosis, Pediatric Research, 1995; 38: 579-584.
to cystine accumulation or to any aspect of his cystinosis . . . Additionally, Arnold-Chiari malformation and persistent fetal circulation complicated his history and suggest other etiologies for the unusual course of his disease. We also cannot blame cysteamine therapy itself; although cysteamine is a duodenal ulcerogen in rats (32), we know of no reports of gastrointestinal dysmotility resulting from the therapy. In fact, the patient's dysmotility preceded the initiation of cysteamine therapy. (Emphasis supplied.)
William A. Gahl, et al., Intravenous Cysteamine Therapy for Nephropathic Cystinosis, Pediatric Research, 1995; 38: 579-584.
The article does not note Shea's Prozac overdose as a significant medical event in Shea's history.
Shea's chronic vomiting and continued intolerance of oral medications, including phosphocysteamine, caused his cystine levels to rise. A course of treatment with phosphocysteamine using the central line was therefore attempted. The article documents the success of that therapy in lowering Shea's cystine levels.
III. DISCUSSION
Defendants do not dispute the scientific reliability of plaintiffs' experts' opinions linking Shea's immediate acute vomiting and nervous system reactions on February 17, 1993 to his inadvertent Prozac overdose. They do, however, attack the scientific validity and reliability of the conclusion that Shea's long term GI decline and resultant inability to continue treatment with cysteamine orally were also caused by his Prozac overdose.
At trial, plaintiff has the burden of showing both specific and general causation. Whitting v. Boston Edison Co., 891 F. Supp. 12, 13 (D.Mass. 1995). In this case, therefore, the Hammonds must show that ingestion of Prozac can cause the types of symptoms alleged in a person with medical conditions similar to Shea; and that the particular dose of Prozac that Shea ingested in this case more likely than not caused his injuries. Defendants' motion in limine seeks to test plaintiffs' proof on both general and specific causation under Daubert and Lanigan, supra.
A. Daubert and Lanigan Standard
Until the Supreme Court's decision in Daubert, federal courts applied the "Frye" standard to test the admissibility of scientific expert opinion. Frye v. United States, 293 F. 1013 (D.C. Cir. 1923). Frye limited such testimony to opinions based on techniques considered "generally accepted as reliable by the relevant scientific community." Daubert at 584. In Daubert, however, the Court abandoned that limitation and focused the inquiry on the relevancy and evidentiary reliability of the expert opinion. It required that the proposed testimony both be "supported by appropriate validation," Daubert at 590, and "be sufficiently tied to the facts of the case that it will aid the jury in resolving a factual dispute." Daubert at 591, (quoting U.S. v. Downing, 753 F.2d 1224, 1242 (3rd. 1985)); In re Paoli RR. Yard PCB Litigation, 35 F.3d 717, 741-43 (3rd Cir. 1994), cert denied, sub nom, General Electric Co. v. Ingram, 115 S.Ct. 1253 (1995). This latter requirement has been termed the "fit" requirement. See e.g. In Re Paoli at 743.
The Massachusetts Supreme Judicial Court adopted these evidentiary standards in Commonwealth v. Lanigan, 419 Mass. 15, 26 (1994). The SJC approved of trial courts taking on a gatekeeping role and making "a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether that reasoning or methodology properly can be applied to the facts in issue." Id. Where there are methodological flaws in the basis of the opinion, including ones in the application of the opinion to the facts or the case, such that the expert lacks good grounds for his or her conclusion, the court should exclude the proffered opinion. Daubert at 2975.
For a proper determination of whether there exist "good grounds," the court should consider whether the reasoning or methodology has been the subject of peer review; whether the theory has been or is able to be replicated; whether the theory is generally accepted as valid within the relevant scientific community; and whether it was developed using sound scientific methods. See Daubert at 593-95; and Lanigan at 25. In some cases this means assessing the validity of the studies upon which the expert depends, See e.g. General Elec, Co. v. Joiner, 118 S.Ct. 512 (1997), in other cases it means assessing whether the stated basis for the opinion is appropriately tied to the facts of the case so that it would help the trier of fact. See Rotman v. National R.R. Passenger Corp., 41 Mass. App. Ct. 317 (1996). In medical causation cases, however, it may mean treading the fine line between the methodology employed and the conclusion reached by the expert. Joiner, supra, 118 S.Ct. at 519. As the Paoli court explained, "a challenge to 'fit' is very close to a challenge to the expert's ultimate conclusion about the particular case, and yet it is part of the judge's admissibility calculus under Daubert." In re Paoli, 35 F.3d at 746.
B. Application of Daubert/Lanigan in the Hammond Case
As noted, at trial, plaintiffs will have to show both general and specific causation. Proof of general causation will require plaintiffs to show that, at the dose ingested, Prozac is capable of causing long term exacerbation of GI symptoms such as those of which Shea Hammond complains. Proof of specific causation will require that plaintiffs show that Shea Hammond's ingestion of Prozac on February 16-17, 1993, more likely than not exacerbated his preexisting GI dysmotility to such an extent that he was unable to continue his cysteamine therapy.
I. General Causation
Dr. Keller makes his opinion based upon his review of medical literature, medical records and depositions from the parties and experts. For his opinion, Dr. Charash depends primarily upon a review of Shea's medical records along with the known adverse side effects of Prozac and the temporal relationship between Shea's symptoms and the overdose. (P. Exhibits 17 and 18.)
The medical literature makes clear that known side effects of regular treatment with Prozac can include nausea, vomiting and various nervous system symptoms. The literature also reveals that the same symptoms are likely in the case of overdose. On its own, however, the literature does not support the conclusion that any potential gastrointestinal and neurological side effects will fail to resolve once treatment with Prozac is discontinued or once the active ingredients in the drug are eliminated from the body.
Dr. Keller does state that "a specific association documented in various publications is that fluoxetine exacerbates a disorder in patients such as Shea Hammond with preexisting disorders and predisposing drug therapy." (Plaintiffs' exhibit 17, p. 8, ¶ 6.) The court, however, has found no support for such a contention in plaintiffs' submissions. Certainly, Shea Hammond presents a unique medical case; he is atypical of the subjects studied in the medical literature. Shea's preexisting condition, in combination with the documented GI side effects of Prozac, however, is insufficient for the court to find that plaintiffs' experts' opinions are scientifically reliable.
The court did find the following similar statement in AHFS at 1699: "[i]n some patients developing movement disorders with fluoxetine, there were underlying risk factors such as predisposing drug therapy and/or the disorder was an exacerbation of a preexisting disorder." (Emphasis supplied.) This statement, however, concerns patients with preexisting movement disorders rather than gastrointestinal ones, and is not applicable to this case.
Of course, an expert opinion developed relying on sources other than peer reviewed literature can be sufficient under Daubert. In re Paoli, supra 35 F.3d at 760. This might include research or evidence accumulated as a result of Dr. Keller's experience, showing a link between Prozac and the exacerbation of GI disorders. It also could include research or other medical literature documenting disorders like those suffered by Shea and the types of events that typically trigger set-backs or relapses in the patient's condition, including ones similar to a drug overdose. Unless plaintiffs' experts show some evidence documenting the links described above, the court concludes that Dr. Keller's expert opinion on general causation is not scientifically reliable enough to reach a jury.
The court comes to the same conclusion regarding Dr. Charash's opinion. His opinion letter discloses little about the methodology he employed when coming to his conclusions. Although Dr. Charash states that he considered the "known" adverse side effects of Prozac including dysphagia, esophagitis, gastritis, glossitis, and stomatitis, the court's review of the literature reveals at least one source stating that a causal relationship between fluoxetine and these diseases has not been established. Like Dr. Keller, Dr. Charash has not provided the court with any additional scientifically valid evidence supporting his opinion.
AHFS at 1700.
ii. Specific Causation
Even if plaintiffs' experts had provided the court with scientifically valid evidence to support general causation, the court also concludes that the methodology they employed in coming to their conclusion that Shea's Prozac overdose in fact caused his long term GI problems is scientifically indefensible under Daubert.
"Courts have insisted time and time again that an expert may not give opinion testimony to a jury regarding specific causation if the expert has not engaged in the process of differential diagnosis." Rutigliano v. Valley Business Forms, 929 F. Supp. 779 (D.N.J. 1996), aff'd, 118 F.3d 1577 (3d. Cir). Indeed, in the Daubert case itself the Supreme Court demanded that the expert "explain how . . . he was able to eliminate all other potential causes of [the] birth defects." Daubert 43 F.3d at 1319. See also In re Paoli R.R. Yard PCB Litigation, 35 F.3d 717, 759 n. 27 and 760 (3rd Cir. 1994). This does not mean that a medical causation expert must eliminate all possible causes of an alleged condition, Claar v. Burlington Northern R.R., 29 F.3d 499, 502 (9th Cir. 1994), nor does it mean that the expert must always make a physical examination of the plaintiff or conduct a battery of tests aimed at eliminating competing causes of the plaintiffs condition. In re Paoli at 762. Nevertheless it is true that "[a]t the core of differential diagnosis is a requirement that experts at least consider alternative causes — this almost has to be true of any technique that tries to find a cause of something." In re Paoli at 759.
The court agrees with the Third Circuit, and concludes that where an expert
engaged in very few standard diagnostic techniques by which doctors normally rule out alternative causes and the doctor offered no good explanation as to why his or her conclusion remained reliable, or the defendants pointed to some likely cause of the plaintiffs illness other than the defendants actions and [the expert] offered no reasonable explanation as to why he or she still believed that the defendants' actions were a substantial factor in bringing about that illness[,]
then the testimony should be excluded as unreliable. In re Paoli at 760.
Plaintiffs have presented the court with the opinion letters of their proposed experts. These letters do not purport to eliminate any other potential cause of Shea's GI problems. Given Shea's complex medical history, including his diagnoses of Cystinosis and GI dysmotility, his Yersinia bacterial infection and his recent brain surgery, there certainly are competing explanations for his GI setback in addition to the Prozac overdose. Shea's medical history reveals repeated instances of severe vomiting and intolerance to feeding. Indeed, the natural course of his GI conditions is one logical explanation for his decline. In this case, the experts each reviewed Shea's medical records. They each also reviewed medical literature regarding the effects of Prozac. As a toxicologist. Dr. Keller has extensive experience in the effects of chemicals on the human body and Dr. Charash is an experienced pediatric neurologist. They each may very well have engaged in the process of considering, evaluating and eliminating other likely causes for Shea's feeding problems. Plaintiffs, however, have not presented evidence to that effect.
In addition, Dr. Charash's opinion relies on the temporal coincidence between Shea's overdose and the failure of his GI system without showing the court the scientific causal link. Temporal coincidence of an event and injury, along with a doctor's experience and observations of patients showing that injury may result from a given event, is insufficient to sustain a plaintiff's burden under Daubert. Rotman at 318. Plaintiffs, therefore, have failed to demonstrate the scientific reliability of the methods that Drs. Keller and Charash used to reach their conclusions.
Finally, the doctors' failure to explain their elimination of alternative causes renders their testimony inadmissible under the "fit" requirement of Daubert. As another court has noted, "expert testimony concerning causation will not assist the trier of fact in reaching an accurate result unless [it] has eliminated alternative explanations for the plaintiffs illness." Rutigliano, supra 929 F. Supp. at 786. See also Rotman v. National R.R. Passenger Corp., 41 Mass. App. Ct. 317 (1996) (without a way to identify MS patients in whom trauma exacerbated optic neuritis, opinion as to cause of optic neuritis was unreliable and would not have assisted jury).
The court is mindful that plaintiffs did not call their experts to testify so that they might demonstrate the methodology they employed in reaching their conclusions; point the court to relevant literature upon which they relied which the court may have overlooked; explain the limitations of literature or studies which might contradict their conclusions; or discuss their own research or experience which might establish the scientific reliability of their conclusions. Given the complex medical conditions involved in this case, in-court testimony may more effectively educate the court than the written submissions presently before it. In light of these considerations, the court will allow plaintiffs the opportunity to present their experts at an evidentiary hearing should they conclude that it would be to their benefit.
ORDER
For the foregoing reasons, defendants' Motion to Exclude Certain Expert Testimony is ALLOWED without prejudice.
Plaintiffs shall notify the court within ten business days of the date below, if they intend to request a hearing to present live testimony on the issues discussed herein.
____________________________ Neel Justice of the Superior Court
Dated: August 17, 1998