The autopsy remains the criterion standard for quality assessment in medicine. Only then are physicians able to see for themselves the true nature of trauma and disease in a person, assess the efficacy of various treatments, and witness the nature of changes within the body as a result of age and lifestyle. Ancillary studies are pivotal in determining the cause and manner of death in many cases. One of the most common ancillary studies is postmortem vitreous chemical analysis, also called vitreous chemistry.
Vitreous fluid is within the globe of the eye, between the retina and the lens. This substance is acellular, viscous, colorless, normally clear, and it is composed predominantly (99%) of water with glucose, hyaluronic acid, collagen fibers (type II), inorganic salts, and ascorbic acid.  Vitreous fluid is ideal for postmortem chemical analysis, as it is relatively isolated from blood and other body fluids that are affected by postmortem changes such as redistribution and hemoconcentration. It also resists putrefaction longer than other body fluids, although it is not entirely immune to it.  In fact, vitreous fluid can be analyzed from bodies that have been previously embalmed (discussed below). Although not common, intrinsic abnormalities or diseases of the eye should be considered when interpreting vitreous fluid results. 
Vitreous Procurement and Pretreatment
This section will discuss procurement and pretreatment of vitreous fluid, including in situations involving embalmed bodies, corneal donation, and child fatalities.
Procurement of vitreous fluid
Vitreous fluid may be procured by inserting an 18- or 20-gauge needle attached to a 10-mL syringe into the globe of the eye. The insertion is best at the lateral canthus, introducing the end of the needle to the center of the globe. The vitreous should be withdrawn slowly. Vacuum tubes should not be used, because the retina can be damaged, thus rendering the specimen inadequate.
Approximately 2-5 mL of fluid can be aspirated from each eye; even 1 mL can be procured from a newborn. The vitreous should be placed in a sterile tube. If alcohol or drug analysis is to be performed, then sodium fluoride preservative should be in the tube. The specimen should be clear and colorless. If small flecks of black-brown retina are in the sample, then the sample is deemed inadequate.
Pretreatment of vitreous
As mentioned earlier, vitreous is viscous. This can present a problem with the instrumentation of the sample. Because the viscosity is largely due to the hyaluronic acid component, hyaluronidase is often used as a liquefying agent before chemical analysis is performed. Alternatively, heating the vitreous at 100°C for 5 minutes followed by cooling is a safe and simple method for improving the precision of the measurement. [4, 5]
Vitreous fluid from embalmed bodies can be analyzed. The fluid is procured as per unembalmed bodies. If the vitreous specimen is pink, it is likely that the embalming fluid has contaminated the vitreous. Embalming fluid contains formaldehyde, glutaraldehyde, perfume aldehydes, EDTA (ethylenediaminetetraacetic acid), germicides, and volatile agents such as methanol and phenol. [6, 7] A separate sample of the embalming fluid should be sent with the vitreous specimen to the laboratory for correlation, if needed. [7, 8]
It is best practice to facilitate tissue donation whenever possible. Close communication with procurement agencies can ensure that they sample vitreous fluid for the pathologist following their sterile procedure.
In cases of abusive head injuries in children (ie, "shaken baby/shaken impact syndrome") removal, fixation, and examination of the eyes is required to rule out or confirm the presence of retinal hemorrhages. In these cases, the pathologist must make a medical determination based on the facts of the case as to sample vitreous fluid or not. Sampling of vitreous fluid may hinder evaluation of the retina in these cases. However, vitreous fluid analysis may be required to confirm coexistent dehydration. The decision to sample vitreous fluid in child and baby fatalities should be made on a case-by-case basis.
Performable Postmortem Vitreous Analyses
Depending on the environment, vitreous fluid can be procured up to approximately 4 days after death. Analyses that can be performed on vitreous fluid include chemistry for the following (see also Table 1 below):
Electrolytes, glucose, ketones
Insulin and C-peptide
Some trace metals
Table 1. Types of Analyses That Can Be Performed on Postmortem Vitreous and Their Applications (Open Table in a new window)
|Sodium||Dehydrations,* water intoxication, low salt pattern, acute ethanol toxicity|
|Potassium||PMI, water intoxication, low salt pattern, acute ethanol toxicity|
|Chloride||Dehydrations,* water intoxication, low salt pattern, acute ethanol toxicity, vomiting,|
|Glucose||Diabetes, nonketotic hyperosmolar coma, DKA|
|Urea (VUN)||Dehydrations,† renal failure, uremia, azotemia|
|Creatinine||Dehydrations,† renal failure, uremia, azotemia|
|Ketones||Fasting/starvation, DKA, alcoholic ketoacidosis, isopropanol ingestion|
|Alcohols||Acute ethanol toxicity, isopropanol ingestion, methanol|
|Drugs‡||Certain drugs can be identified, quantified, and correlated with blood concentrations|
|Antibodies||Viral antibodies to HIV and hepatitis B and C in donation cases|
|6-MAM||To distinguish heroin from morphine|
|DNA||For purposes of identification|
|Formic Acid||Confirm premortem methanol ingestion|
DKA = diabetic ketoacidosis; DNA = deoxyribonucleic acid; HIV = human immunodeficiency virus; 6-MAM = 6-monoacetylmorphine; PMI = postmortem interval; VUN = urea nitrogen concentration in vitreous humor.
* Hypernatremic and hyponatremic dehydrations.
† Urea nitrogen and creatinine are useful in assessing hypernatremic, isonatremic, and hyponatremic dehydrations.
‡ Cocaine, heroin, morphine, tricyclic antidepressants, barbiturates, benzodiazepines, and gamma-hydroxybutyrate (GHB) are some of the drugs that can be identified in vitreous fluid.
Table 2 summarizes the findings of vitreous analyses in some common conditions.
Table 2. Postmortem Vitreous Chemical Analyses in Some Common Diseases (Open Table in a new window)
|Condition||Na mmol/L||Cl mmol/L||K mmol/L||Cr mg/dL||VUN mg/dL||Glucose mg/dL||Ketoacids (Pos or Neg)||R-OH mg/dL|
|Normal Vitreous||135-150||105-135||< 15||0.6-1.3||8-20||< 200||Neg||Neg|
|Hyponatremic Dehydration||< 135||< 105||Maybe increased||Increased|
|Low Salt Pattern||< 135||< 105||< 15|
|Water Intoxication||< 135||< 105||< 15|
|Decomposition||< 130||< 105||>20||Pos|
|Nonketotic Hyperosmolar Coma||Maybe decreased||Maybe increased||Maybe increased||>200|
|Alcoholic Ketoacidosis||< 200||Pos|
|Acute Ethanol Toxicity Binge||< 135||< 105||< 15||>350|
Table adapted from Rose KL, Collins KA. Vitreous postmortem chemical analysis. NewsPath. December 2008.
Cl = chloride; Cr = creatinine; K = potassium; Na = sodium; Neg = negative; Pos = positive; R-OH = alcohol; SIADH = syndrome of inappropriate secretion of antidiuretic hormone; VUN = urea nitrogen concentration in vitreous humor.
Electrolytes: sodium, potassium, chloride, urea nitrogen, creatinine
After death, cell membranes become permeable. Active and selective membrane transport stops, and the loss of selective membrane permeability and diffusion of ions, and other parameters according to their concentration gradients, starts.  Although stable for longer time periods postmortem, certain vitreous elements will change with diffusion from the retinal cells. Potassium immediately begins to diffuse out of these cells into the vitreous, and the level of potassium increases linearly. For this reason, potassium has been used to estimate postmortem interval. [11, 12] However, the reliability of estimated postmortem interval decreases with time and increasing potassium concentration. [4, 13, 14] Furthermore, postmortem studies have shown that vitreous potassium levels may vary between the eyes of the same individual at the same time. Lastly, factors that accelerate decomposition (eg, hyperpyrexia) can also affect the slope of the vitreous potassium increase.
Sodium, chloride, creatinine, and urea nitrogen are more reflective of the premortem blood levels at the time of death and can useful in diagnosing diseases. Diseases with electrolyte imbalances include dehydration, water intoxication, forced salt ingestion, burns, diuretic use, Addison disease, syndrome of inappropriate secretion of antidiuretic hormone (SIADH), renal failure, uremia, azotemia, and gastroenteritis with vomiting and/or diarrhea. [15, 16, 17, 18, 19, 20] The environmental temperature, postmortem interval, nutritional status, decedent's age, and agonal period must be considered when attempting to interpret electrolyte levels and apply relevance to the results.
Vitreous glucose decreases immediately after death. Therefore, hypoglycemia is extremely difficult to diagnose by postmortem glucose levels. A level of 180 mg/dL or less is considered normal. A level greater than 200 mg/dL is indicative of diabetes mellitus, diabetic ketoacidosis, or nonketotic hyperosmolar coma.  levels less than 20 mg/dL are commonly encountered and are of little diagnostic significance.
Normally, ketones are not in the vitreous. Ketones that can be detected in certain diseases include acetone, acetoacetate, and beta-hydroxybutyrate. Beta-hydroxybutyrate is the major ketone produced in diabetic ketoacidosis and alcoholic ketoacidosis. [22, 23, 24, 25] Vitreous acetone can be detected in cases of starvation, extreme dieting, fasting, or malnutrition. Acetone production secondary to a low caloric intake can result in the conversion to low levels of isopropanol. However, acetone can be produced when an individual ingests isopropanol. The reaction between acetone and isopropanol is a reversible one. The isopropanol/acetone ratio will help in determining the underlying pathology.
As mentioned previously, isopropanol may be detected after ingestion of or from the conversion of acetone to isopropanol. The isopropanol level will be higher when ingested as opposed to conversion from acetone. Methanol and its toxic metabolite formic acid can be detected after accidental or intentional ingestion. [26, 27, 28]
Ethylene glycol and metabolites can also be identified in postmortem vitreous.  After ethanol ingestion, ethanol can be in the vitreous fluid approximately 2 hours after oral ingestion. Ethanol can also be produced by microbial activity and the fermentation of glucose postmortem. [30, 31, 32, 33, 34, 35] In fact, up to 100 mg/dL may be produced by bacteria alone in cases of decomposition. Postmortem diffusion of ethanol from the stomach to the central blood is an important factor to consider when interpreting postmortem blood ethanol levels. [2, 30, 31]
Vitreous fluid is anatomically removed from the large blood vessels and the gastrointestinal organs and thus is a choice specimen. Vitreous fluid is useful in cases of drowning due to possible dilution of other body fluids, decomposition, and postmortem synthesis of ethanol.  The ethanol level in the vitreous corroborates with premortem ethanol use; however, interpretation can be difficult due to an approximate 1-2 hour lag time and whether the individual is in the absorptive or postabsorptive/metabolic phase. If the vitreous ethanol level is lower than the blood level, the individual was still absorbing ethanol at the time of death. If the vitreous ethanol level is higher than the blood level, the individual was in the metabolic phase and the blood level had been higher previously. The vitreous ethanol level will be lower than the blood level only when in the absorptive phase. [33, 34, 35]
It is possible that every drug that is detected in the blood can be detected in the vitreous fluid. The vitreous levels of certain drugs have been studied and are important for their identification as well as correlation with premortem levels. However, the significance of a vitreous level may be of question. For example, postmortem diffusion of certain drugs from the brain into the vitreous may occur.  Also, hydrophilic drugs are more likely to have concentrations approaching those of blood or plasma than those that are highly protein-bound (such as tricyclic antidepressants [TCAs]) or lipophilic (such as benzodiazepines).
Alcohols (see above), cocaine and its metabolite benzoylecgonine, morphine, heroin and its metabolite 6-monoacetylmorphine (6-MAM), gamma-hydroxybutyrate (GHB), and TCAs are some of the drugs that have been identified. The utility of cocaine detection is limited to the presence of the drug and, in fact, it has been reported that the vitreous concentration of cocaine can increase with time. [33, 34, 35, 36]
Diacetylmorphine (heroin), morphine, and 6-MAM can all be detected in the vitreous fluid. The presence of 6-MAM is useful in determining that morphine was derived from heroin, as opposed to the ingestion of parent morphine.  GHB can be found as an endogenous neurotransmitter, recreational drug, or therapeutic agent. Postmortem production of GHB can occur in the blood. Therefore, vitreous analysis is good for confirming exogenous GHB. [38, 39]
Antibodies to the human immunodeficiency virus (HIV) can be detected in the vitreous. This has potential usefulness in cases of cornea donation. 
Stress hormones: catecholamines
Stress hormones and noradrenaline have been analyzed to correlate with the length of the agonal period. The vitreous presence and levels have been studied in cases of hypothermia, asphyxia, and cardiopulmonary resuscitation. The hormones can be quantified but, thus far, the levels do not correlate with the length of the agonal period in individual cases.  (Asphyxia will be discussed in a separate article.)
Acids and acid metabolites can be detected in the blood and vitreous fluid in a large variety of conditions and diseases. Lactic acid can be detected in the vitreous in cases of hyperglycemia and is formed during the anaerobic metabolism of glucose. [7, 20] Some researchers have assessed lactate and glucose levels and the sum total to diagnose diabetes mellitus. However, lactate levels increase linearly after death, and this sum total can be misleading, resulting in a false diagnosis of diabetes mellitus. 
Lactic acid is also produced in cases of diabetic ketoacidosis, alcoholic ketoacidosis, cyanide ingestion, renal and liver disease, and iron toxicity. Methanol metabolism produces formic acid. Ingestion and metabolism of ethylene glycol result in the formation of lactic acid, formic acid, glycolic acid, and oxalic acid. Propylene glycol is broken down to lactic acid, acetic acid, and pyruvic acids. In cases of acetylsalicylic acid (aspirin) (ASA) toxicity, the parent as well as acid metabolites can be detected. Elevated amino acids and organic acids in the vitreous can be a sign of an inborn error of metabolism.
Insulin and C-peptide
Proinsulin is enzymatically cleaved to insulin and C-peptide. Insulin and C-peptide are secreted in equimolar concentrations, and both can be detected and quantified in the vitreous in cases of insulin overdoses. Of note, C peptide is not present in exogenous insulin, human or animal. Therefore, a high level of insulin and a low level of C-peptide is indicative of an exogenous insulin overdose. 
Metals can be detected in the vitreous fluid. Iron toxicity is one such scenario. Vitreous magnesium levels have also been studied. Alcoholic individuals can have a decreased serum magnesium premortem; however, no significant difference in levels has been detected between alcoholic and nonalcoholic persons.  Vitreous metal concentrations can also be affected by heat, such as in fires, which can cause an increased rate of intracellular release. 
Due to the previously mentioned qualities of vitreous fluid, several other interesting, although currently inconclusive, studies have been reported. Serum proteins such as the amino terminal of probrain natriuretic peptide (NT-proBNP) as a discerning marker of heart failure and early cardiac dysfunction show potential.  The analysis of apolipoprotein (APO) levels in the vitreous of diabetics may correlate with severity of the disease. 
In addition, combining the concentrations of potassium, urea, and hypoxanthine with varying methodologies has been utilized to better estimate postmortem interval.  Even the analyses of carbonylated proteins, markers of oxidative stress, have been correlated with the biologic age of an individual. 
Vitreous fluid potassium levels cannot pinpoint the time of death. Although it is currently the best biochemical marker for determining the postmortem interval, vitreous fluid potassium levels are subject to many variables. In fact, most medical examiners do not rely on the vitreous potassium level to determine the time of death, despite the popularity of this technique in fictional adaptations of forensic practice.
Hypoglycemia cannot be confirmed or ruled out based on vitreous fluid analysis.
Although a markedly elevated vitreous glucose level can indicate hyperglycemia, a normal vitreous glucose concentration cannot exclude it.
Vitreous is a unique body fluid: it is unique in its composition, its anatomic location, and its isolation from other body fluids. Vitreous composition reflects serum concentrations of many elements in the immediate pre-mortem period. It is easily retrievable and simple to analyze. In addition, many exogenous molecules and chemicals can be identified in the vitreous, and many diseases are reflected in its composition. As researchers continue to focus on the postmortem analysis of vitreous fluid, more applications, precise interpretations, as well as limitations will be discovered.