The Death of SIDS

Abstract

Sudden Infant Death Syndrome (SIDS) is an overly broad classification bin for sudden unexplained infant deaths. SIDS has become a “diagnostic” phrase that encompasses unidentified, disease-related causes of death, deaths likely due to accidental asphyxia, and possibly unrecognized homicides. There is a prevailing false concept that SIDS is a “real” and discrete diagnostic entity rather than a phrase that signifies an inability to state why an infant has died. This has been perpetuated by the International Statistical Classification of Diseases and Related Health Problems (ICD-10), which recognizes SIDS as a “cause” of death. We propose simplified, unambiguous language for the death certificate in cases of sudden unexplained infant death. We propose changes to ICD-10 nomenclature and vital statistics tabulation practices of the National Center for Health Statistics (a division of the Centers for Disease Control and Prevention) to end the use of SIDS as a diagnosis.

Keywords

Cause of death Death certification Sudden unexpected infant death

Introduction

The entrenched lexicon of sudden unexplained infant death, long dominated by Sudden Infant Death Syndrome (SIDS), is changing. Originally created in 1969 to provide a classification mechanism to group and study unexplained infant deaths, SIDS evolved from a descriptor into a diagnosis as if it were a singular disease or disorder with an as yet undiscovered cause. Decades of research have largely focused on potential “causes” of SIDS from a disease-based pathobiological standpoint rather than focusing on a deliberate synthesis of autopsy and death scene information.

We present an overview of SIDS definitions, the nosology (vital statistics classification of mortality data) of SIDS in the United States, and the diagnostic shift that has occurred as a result of improved death scene investigation. Additionally, we present a death investigation/forensic pathology view of sudden unexpected infant death, particularly as it pertains to rational conclusions regarding 1) determining cause and manner of death in the forensic setting and 2) the language of death certificates. Forensic pathologists, the physicians responsible for investigation of sudden unexpected infant deaths, are propagating migration away from the use of “SIDS” as a cause of death. As with novel diseases and evolving understanding of known diseases, nomenclature must adapt. SIDS is a “disease” name that has run its course, to be replaced in our vocabulary by “undetermined”, an honest admission of “I don't know.”

History of Sids and Transition to Suid

SIDS has a 40+ year history as a definitional tool to “diagnose” and study sudden unexplained infant death. SIDS - a diagnosis of exclusion - is an anchored fixture of the scientific and lay concepts of sudden unexplained infant death.

The “Sudden Infant Death Syndrome” was codified in 1969 by a National Institutes of Health consensus committee (1). Panelists recognized that many unexplained infant deaths shared common features (2-4 months old, lower socioeconomic level, seasonality, death while asleep) but a single explanatory disease or mechanism of death was not known. Autopsies were unrevealing aside from variable intrathoracic petechiae and variable mild inflammation of the respiratory tract, which are now generally accepted to be non-specific.

The committee's enduring product (20 years without an update) was this definition of SIDS: “The sudden death of any infant or young child, which is unexpected by history, and in which a thorough post-mortem examination fails to reveal an adequate cause for death” (1). Of note, the importance of death scene investigation and the clinical history, while assumed to be obvious, was not explicit in the 1969 definition.

A guiding committee objective was “to focus attention and research activity on this problem.” Practical outcomes were to 1) name this group of cases and 2) provide terminology for the cause of death field on death certificates (DCs) (2). And so began the practice of using the single moniker SIDS to define this heterogeneous group of unknown causes of death.

Modest updates in 1989 and 2004 defined the age range, stated the importance of scene investigation and the clinical history, and recognized the temporal relationship with sleep (3–4). In 2005, the National Association of Medical Examiners (NAME) ad hoc Committee on Sudden Unexplained Infant Death proposed new nomenclature, invoking the committee name as improved language for cases that would be labeled SIDS under earlier definitions. Sudden unexplained infant death (SUID) was defined as: “The death of an infant less than one year of age in which investigation, autopsy, medical history review and appropriate laboratory testing fail to identify a specific cause of death. SUID includes cases that meet the definition of sudden infant death syndrome” (5). The 2005 definition was essentially identical to the 2004 definition except that the syndromic name was replaced by a descriptive statement.

The committee recommended using “Sudden Unexplained Infant Death” for the cause of death statement on the DC. Unfortunately, in order to account for practice variation among forensic pathologists, the committee wavered and listed other cause of death statements they considered synonymous with SUID and acceptable on a DC. These are “sudden death during infancy: no identifiable cause”, “consistent with the definition of Sudden Infant Death Syndrome”, “consistent with Sudden Infant Death Syndrome”, and “Sudden Infant Death Syndrome.” Acceptance of these synonyms allowed continued non-standardized language on DCs based on personal preference and office or regional tradition, thereby diluting the effectiveness of the recommendation as a tool for change.

The committee suggested attaching risk factors or stressors (e.g. bedsharing and high environmental temperature) to the DC entry in either Part I (cause of death) or Part II (significant conditions contributing to death). The committee recommended classifying all SUID deaths as undetermined manner of death, an abrupt transition from the prior natural manner convention that had gained variable acceptance within the forensic pathology community.

The 2005 definition was a paradigm shift in three ways. First, the descriptive cause of death statement, SUID, could not be misused as an actual diagnosis. Second, potential risk factors integrated into the DC could communicate modifiable behaviors. And third, an undetermined manner of death acknowledged an asphyxial or other nonnatural component of some of these deaths. If the cause of an infant death is unknown, it is generally not reasonable to conclude that the manner of death is natural.

Nosology of Sudden Unexplained Infant Death

Past and current nosologic practice in the United States (US) is SIDS-centered. Coding rules do not capture the spectrum of death certificate language in sudden unexplained infant deaths, and coding rules have less current relevance since many certifiers have abandoned SIDS as a diagnostic phrase.

The fundamental objective of mortality coding is to facilitate the tabulation of cause of death statistics abstracted from the many diseases and/or injuries entered on DCs. Coding of the underlying cause of death adheres to the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10), implemented in the US in 1999. Coding is largely automated by software designed to apply the ICD coding rules. The software translates free text COD information on DCs to ICD-10 codes and selects the underlying COD. For tabulation of aggregate underlying COD data, ICD-10 codes are merged into a national database managed by the National Center for Health Statistics (NCHS) of the Centers for Disease Control and Prevention (CDC) (6). The NCHS is required to use ICD as its classification and coding system.

Deaths and death rates are reported in annual National Vital Statistics Reports according to a standard list of underlying cause-of-death categories. Among some 2.42 million deaths in the US in 2007, the most common categories of underlying COD were “diseases of heart”, “malignant neoplasms” and “cerebrovascular diseases”, in that order. Of note, these data blend cause and manner of death information: accidents, suicides and homicides occupied the 5th, 11th and 15th positions for leading “causes” of death, respectively, although these are manners of death (circumstances) rather than causes of death (6).

Among some 29,000 infant deaths in 2007, leading causes of death were “congenital malformations, deformations and chromosomal abnormalities”, “disorders related to short gestation and low birth weight”, and “Sudden Infant Death Syndrome (SIDS)”, respectively. The total number of SIDS deaths was 2,453. Accidental death was the 5th leading “cause” of death in this cohort (6).

Rules that govern COD coding recognize “ill-defined and unknown causes” of death. These include SIDS (coded as R95) and unknown cause (R99). Many diverse DC entries are coded as SIDS, e.g. “Cot Death”, “Crib Death”, “Sudden Death in Infancy”, “Sudden Infant Death”, “Sudden Unexpected Death”, “Sudden Unexplained Death of an Infant”, and “Sudden Unexpected Infant Death” (SUID). All are collapsed into the R95 SIDS code. Similarly, “unknown cause” may be written by a certifier in many ways, e.g. “cause not found”, “etiology not defined”, “no specific known causes”, “uncertain”, and “unclear”. All are coded as R99 (7).

Many medical examiners and coroners who complete death certificates are unaware of the coding rules that funnel unexplained infant deaths with disparate case details into SIDS R95 or unknown cause R99. Whatever the content of an individual DC for an unexplained infant death, the distilled data are limited to R95 or R99. Also not commonly understood is that other text that lists risk factors or environmental stressors, which some medical examiners/coroners include on the DC to provide additional rationale or explanation, is neither collected nor coded in a systematic way.

Prevalence of SIDS and Diagnostic Shift to Asphyxia and Undetermined/Unknown

United States vital statistics demonstrate trends of 1) reduced classification of deaths as SIDS (R95), 2) increased diagnosis of infant asphyxial deaths, and 3) increased number of infant deaths classified as unknown cause of death (R99).

According to ICD coding rules, the number of infants who “die of SIDS” (R95) is the sum of all DCs that contain SIDS, SUID or similar language. Some infant DCs are explicit about the absence of an identifiable cause of death (e.g. “undetermined cause of death”). Although SIDS and SUID are by definition unexplained, DCs that explicitly indicate an unidentified COD are coded separately as unknown cause (R99). Asphyxia is a discrete injury process and, when entered as COD on a DC, is coded as W75.

Malloy and MacDorman, and Shapiro-Mendoza et al. analyzed national mortality data through 2001. This was the decade following the 1992 American Academy of Pediatrics recommendation favoring the supine sleep position for infants (the “Back to Sleep” campaign started in 1994) (8–9). The all-cause postneonatal death rate declined at an average annual rate of 3.8%. R95 SIDS deaths declined at an annual rate of 8.6%, from 4,593 to 2,047. Deaths attributed to asphyxia or R99 unknown cause increased annually over the 1992-2001 time period. The annual number of infant homicides was stable.

These mortality trends occurred during a time of significant change in infant sleep position. Approximately 70% of infants were put to sleep prone in 1992; the prone rate was approximately 20% by 1997 and had stabilized in the mid to low teens in the decade starting with 2000. The prone put to sleep position was 11.4% in 2009 (10).

The declining all-cause postneonatal death rate stabilized during the last three years (1999-2001) of the studies’ time period and corresponded to a stabilization of the now dominant supine sleep position. This suggested that the impact of Back to Sleep had ended. During the same three years, the SIDS classification rate continued to decline and the asphyxia and unknown/unspecified rates continued to increase. Authors of both studies cautiously proposed that some deaths that would have previously been attributed to SIDS on a DC were instead attributed to asphyxia or to an unknown cause. An additional focused review of the infant death rate due to asphyxia, through 2004, revealed a continuing increase in the diagnosis of infant asphyxial deaths (11).

This diagnostic shift was due, at least in part, to incorporating scene investigation into the previously autopsy-based culture of infant death investigation. Rigorous scene investigation is now standard practice in investigation of sudden unexpected infant death in the US. Retrospective studies from the Wayne County, Michigan (Detroit) and Maryland State Medical Examiner Offices starkly demonstrate changes in death certification practice associated with detailed scene investigation. In both jurisdictions, there was a marked reduction in use of SIDS as cause of death offset by a marked increase in deaths attributed to asphyxia or an unknown cause (12–13). Remarkably, the Michigan study demonstrated one or more exogenous risk factors for asphyxia (overlying or entrapment, bed/couch sharing, sleep position with obstruction of the nose and mouth, soft bedding, coverage by bedding) in 85% of cases (12).

It must be reinforced that these exogenous factors are environmental characteristics that are risk factors for death by an asphyxial mechanism. These include rebreathing of exhaled air, accidental smothering due to prone sleep position or entanglement in bedding, and overlaying by bed/couchsharing adults or children (14–15). Studies of diagnostic shift indicate that thoughtful interpretation of every infant's environment around the time of death resulted in more deaths being classified as asphyxia in nature.

Discussion

The 1969 SIDS concept brought definition to disparate, perplexing cases of sudden unexplained infant death. There were unintended consequences.

•

First, SIDS was intended as a category of exclusion but became a diagnosis of exclusion.

•

Second, SIDS as used by physicians seeing family members of a deceased infant and as used by physicians completing DCs implied that death was due to an unidentified disease process.

•

Third, by convention the manner of death noted on SIDS death certificates was “natural.” The presumption of disease causation minimized the possibility of an environmental contribution to death or an environmental cause of death.

•

Fourth, the ICD and, by extension, NCHS became fully wedded to the concept that SIDS was a cause of death.

It is inaccurate and misleading to state that an infant “died of SIDS.” This may appear to be an argument of semantics, but consider how SIDS has evolved from a descriptive classification (fundamentally, an adjective) to a disease (a noun). Infants do not die of a classification system. This misuse of the original intent of SIDS has widely penetrated medical research, the medical literature, the culture of law enforcement, the common lay concept of infant death and medical certification of cause of death.

To this end, many pediatricians, parents, law enforcement officers and even some forensic pathologists presuppose SIDS as the most likely “cause” of death when first presented with a case of sudden unexpected infant death without visible external trauma. This perspective may be self-fulfilling in the context of a negative autopsy. The differential diagnosis of sudden unexpected infant death includes natural diseases or anatomic abnormalities (e.g. myocarditis or Tetralogy of Fallot), cardiac channelopathies (16), intrinsic and extrinsic conditions (e.g. extreme prematurity, hyperthermia, or malnutrition), and injuries (e.g. blunt force trauma or traumatic asphyxia, possibly including undetected homicides). SIDS is not a cause of death and cannot be in the differential.

Knowledge of sleep position and environment is now an essential component of each case analysis. Excellent scene investigation and a thorough autopsy are interdependent in every death investigation, and especially so in sudden unexpected infant deaths. It is now recognized that many infant deaths previously classified as SIDS are deaths due to an extrinsic environmental injury (e.g. asphyxia) or combined intrinsic/extrinsic mechanisms (e.g. possible impaired arousal reflex triggered by environmentally-induced hypoxia) (17).

The Back to Sleep campaign and enhanced investigative description of a deceased infant's sleep environment, in particular, have been temporally associated with a diagnostic shift in cases without a cause of death revealed by autopsy (a “negative” autopsy). Some infant deaths are unequivocal accidental asphyxial deaths based on compelling scene information and death investigators are increasingly willing to interpret these as asphyxial deaths. Some contend that this observed trend is overaggressive interpretation of a sleep environment as unsafe. It is important to realize, however, that the quality of death investigation has improved exponentially over the past four decades and, if anything, asphyxial deaths were underdiagnosed in the past rather than overdiagnosed presently.

Death investigators are increasingly adept at recognizing a grey zone of possible asphyxial deaths based on environmental risk factors but without definite overlaying or entrapment. Current COD terminology on DCs for these grey zone cases runs the gamut from “SIDS” to “possible asphyxia” to “sudden unexplained infant death” to “undetermined”, among others.

Absence of asphyxial or other external risk factors, in the setting of a negative autopsy, presents another diagnostic challenge. COD terminology for these cases ranges from “SIDS” to “SUID” to “consistent with SIDS” to “undetermined”, and a host of others. Aside from “asphyxia” (and smothering, etc.) and “undetermined” causes of death, all of the above listed causes of death are classified as R95 SIDS per coding rules. Obviously, this is often not the intent of the medical examiner or coroner who entered cause of death information on the death certificate. Even though many death certifiers are intentionally trying to avoid the term “SIDS”, these cases are still being coding as such.

Some death certifiers will routinely list potential risk factors in the COD statement, either in the “other significant conditions” or the “how injury occurred” sections of the DC. While the intent of the certifier in doing so is understood, it is procedurally inappropriate and perpetuates the belief by some that medical certifiers are overzealous in contributing accidental risk factors to the cause of death. These DC fields are not intended to provide a rationale for indicating a possible cause of death when the certifier has already certified the cause as essentially “undetermined”.

Is there value in a potpourri of generally synonymous diagnostic terms, all of which attempt to convey an undetermined cause of death but are more or less derived from SIDS? There is no nosologic value as indicated by ICD coding rules, which translate many causes of death into either R95 SIDS or R99 unknown regardless of what the certifier intended. There is no value to parents and, on the contrary, diverse SIDS equivalent phrases may cause confusion.

Recommendations

For cases of sudden unexpected infant death in which the COD is not known, the medical examiner/coroner should enter “undetermined” in the cause and manner sections of the death certificate. Do not enter risk factors for sudden unexpected infant death on the DC.

We propose a return to simplicity and transparency of meaning, thus “calling it what it is”. The Sudden Infant Death Syndrome classification was created to capture infant deaths of undetermined cause. Let us extract this core definition of SIDS as a word – the word – that best states the final summary and conclusion of infant death investigations that do not yield a cause of death. These DCs should contain one word in the COD section: undetermined. As recommended by Corey et al. in the 2005 definition, the manner of death should also be classified as undetermined (5). Death investigators are well accustomed to invoking “undetermined” for cause and manner of death in every other age group – we should extend this practice to infants. Not surprisingly, “Sudden Elder Death Syndrome” (SEDS) has been invoked as a spoof by The Onion, a satirical online newspaper (18).

SIDS was also preferred as a “gentler” classification term to allay feelings of guilt on the part of parents, presuming that “unexplained” imparted an “aura of mystery” (1). We disagree with this reasoning. Parents, of all people, should be entrusted with a clearly stated, honest and forthright conclusion, even if that conclusion lacks the solidity of a specific diagnosis such as pneumonia or congenital heart disease. Good communication with parents should ensure an adequate explanation that “undetermined” simply means “unable to be determined” or “we don't know”. Undetermined does not necessarily imply that the death is “suspicious.” Further, designating the cause of death as undetermined should not diminish the parents’ access to appropriate grief counseling.

Epidemiologic studies have identified dangerous or risky sleep environments and sleep positions. There is no nosologic value in listing these variables on the death certificate because free text information is not coded by the NCHS for aggregate analysis of COD data. Observations or impressions regarding risk factors or other specific case data are best summarized in the individual autopsy report and the accompanying detailed, standardized investigation report. The content of these reports will be collated by any competent child death review team. Death investigators who ultimately certify these infant deaths are obligated to communicate with families, public health agencies and legal agencies in a clear, succinct fashion – the death certificate is not the appropriate or useful medium for discussing risk factors that may or may not qualify as contributing conditions or for discussing other case details (minor pulmonary infection, remote rib fracture, etc.) of uncertain significance.

Revise ICD coding rules in ICD-11 to eliminate R95. Fold all undetermined causes of death, including SIDS and SUID if medical examiners continue to use these terms on death certificates, into unknown cause of death R99. Alternatively although less optimal, revise NCHS tabulation lists to eliminate SIDS as a discrete cause of death category and add deaths coded as R95 to R99 deaths to yield a sum total infant deaths of unknown cause of death in official tabulations and rankings.

SIDS (R95) has outlived its usefulness as a tool of epidemiology and research. Recall that the original SIDS definition was conceived to “to focus attention and research activity on this problem.” This was successful – the 1969 definition and subsequent revisions brought about decades of improved case capture, abundant research, and a global public health effort in the form of the Back to Sleep campaign. The unintended negative consequence was to create a new “disease” that grossly mischaracterized the diversity of sudden unexpected infant death.

The recent trends in cause of death certification are reduced prevalence of SIDS classification, increased prevalence of asphyxia death classification, and increased prevalence of undetermined cause of death classification. These aggregate changes have occurred with improved infant death investigation, particularly scene investigation. Many MEs and coroners now have the investigative tools to identify other causes of infant deaths such as asphyxia or the long QT syndrome. Many now do not invoke SIDS due to a preference for more transparent certification of these unexplained deaths. ICD and NCHS coding and tabulation practices should reflect these current practices of death certification.

Conclusion

SIDS deaths are infant deaths of undetermined cause. The SIDS umbrella was a powerful, unifying research tool during the late 20th century to facilitate research and public health. That umbrella is now frayed, with diminished utility to capture and categorize sudden unexplained infant deaths.

Improved certification of sudden unexpected infant deaths is based on compulsive scene investigations, detailed reviews of medical and family histories, excellent autopsies and consistent performance of a spectrum of associated studies. More cases are now identified as asphyxial deaths; many asphyxia cases have been misclas- sified as SIDS for decades. Many deaths remain unexplained, and some of those are likely to represent several different natural diseases or conditions that we are currently unable to identify. Research should and must continue in the area of unexplained infant death, but not in SIDS per se.

In the absence of an identifiable cause and manner of death, the unexplained and unexpected death of an infant should be classified as undetermined for both the cause and manner of death on death certificates and in national vital statistics. ICD definitions and coding rules and NCHS tabulation lists should adapt to evolution of disease and injury nomenclature. While updates cannot be undertaken lightly, a sea change in diagnostic language (as seen with sudden unexpected infant death) should be followed by parallel changes in nosology practice.

Footnotes

Acknowledgements

We thank Drs. Stephen Cina, Craig Mallak, and Erin Presnell for helpful discussions and critical review of the manuscript. We thank Dr. Robert Anderson for guidance in the discussion of death certification nosology.

References

Bergman

A.B.

, Beckwith

J.B.

, Ray

C.G.

, editors. Sudden Infant Death Syndrome; Proceedings of the Second International Conference on Causes of Sudden Death in Infants. Seattle: University of Washington Press; 1970.

Beckwith

J.B.

. Defining the sudden infant death syndrome. Arch Pediatr Adolesc Med 2003 Mar; 157(3): 286–90.

Willinger

, James

L.S.

, Catz

. Defining the sudden infant death syndrome (SIDS): deliberations of an expert panel convened by the National Institute of Child Health and Human Development. Pediatr Pathol 1991 Sep-Oct; 11(5): 677–84.

Krous

H.F.

, Beckwith

J.B.

, Byard

R.W.

, Rognum

T.O.

, Bajanowski

, Corey

. Sudden infant death syndrome and unclassified sudden infant deaths: a definitional and diagnostic approach. Pediatrics 2004 Jul; 114(1): 234–8.

Corey

T.S.

, Hanzlick

, Howard

, Nelson

, Krous

; NAME Ad Hoc Committee on Sudden Unexplained Infant Death. A functional approach to sudden unexplained infant deaths. Am J Forensic Med Pathol 2007 Sep; 28(3): 271–7.

, Kochanek

K.D.

, Murphy

S.L.

, Tejada-Vera

. Deaths: Final Data for 2007. Natl Vital Stat Rep [Internet]. 2010 May 20; 58(19): 1–135. Available from: http://www.cdc.gov/nchs/data/nvsr/nvsr58/nvsr58_19.pdf.

National Center for Health Statistics (U.S.). Instruction manual. Part 2a, Instructions for classifying the underlying cause of death. 2010 [cited 2010 March 3]; Available from: http://www.cdc.gov/nchs/data/dvs/2A_2010acc.pdf.

Malloy

M.H.

, MacDorman

. Changes in the classification of sudden unexpected infant deaths: United States, 1992-2001. Pediatrics 2005 May; 115(5): 1247–53.

Shapiro-Mendoza

C.K.

, Tomashek

K.M.

, Anderson

R.N.

, Wingo

. Recent national trends in sudden, unexpected infant deaths: more evidence supporting a change in classification or reporting. Am J Epidemiol 2006 Apr 15; 163(8): 762–9.

10.

National Infant Sleep Position. Data Coordinating Center, Boston University School of Public Health; 2004 [updated Oct. 21, 2009; cited 2010 March 3]; Available from: http://dcc2.bumc.bu.edu/Chimenisp/NISP_Data.asp

11.

Shapiro-Mendoza

C.K.

, Kimball

, Tomashek

K.M.

, Anderson

R.N.

, Blanding

. US infant mortality trends attributable to accidental suffocation and strangulation in bed from 1984 through 2004: are rates increasing? Pediatrics 2009 Feb; 123(2): 533–9.

12.

Pasquale-Styles

M.A.

, Tackitt

P.L.

, Schmidt

C.J.

. Infant death scene investigation and the assessment of potential risk factors for asphyxia: a review of 209 sudden unexpected infant deaths. J Forensic Sci 2007 Jul; 52(4): 924–9.

13.

, Zhang

, Zielke

R.H.

, Ping

, Fowler

D.R.

. Observations on increased accidental asphyxia deaths in infancy while cosleeping in the state of Maryland. Am J Forensic Med Pathol 2009 Dec; 30(4): 318–21.

14.

Kemp

J.S.

, Kowalski

R.M.

, Burch

P.M.

, Graham

M.A.

, Thach

B.T.

. Unintentional suffocation by rebreathing: a death scene and physiologic investigation of a possible cause of sudden infant death. J Pediatr 1993 Jun; 122(6): 874–80.

15.

Kemp

J.S.

, Thach

B.T.

. Quantifying the potential of infant bedding to limit CO2 dispersal and factors affecting rebreathing in bedding. J Appl Physiol 1995 Feb; 78(2): 740–5.

16.

Van Norstrand

D.W.

, Ackerman

M.J.

. Sudden infant death syndrome: do ion channels play a role? Heart Rhythm 2009 Feb; 6(2): 272–8.

17.

Kinney

H.C.

, Thach

B.T.

. The sudden infant death syndrome. N Engl J Med 2009 Aug 20; 361(8): 795–805.

18.

Millions of Americans succumbing to Sudden Elder Death Syndrome. Madison, WI: The Onion[Internet]. 2006 Feb. 8[cited 2010 March 7];(42-06): 1. Available from: http://www.theonion.com/content/node/45109.