Minimally invasive thoracic corpectomy and anterior fusion in a patient with metastatic disease: case report and review of the literature

For patients with metastatic disease to the spine there are numerous surgical approaches for decompression of neural elements and maintenance of mechanical stability. The challenge is to accomplish this while minimizing patient morbidity. Here we report on the feasibility and utility of a minimally invasive extreme lateral approach to the mid to high thoracic spine for anterior decompression and fusion.

Apolipoproteins, cardiovascular risk and statin response in type 2 diabetes: the Collaborative Atorvastatin Diabetes Study (CARDS)

AIMS/HYPOTHESIS

Controversy surrounds whether the ratio of apolipoprotein B (ApoB) to apolipoprotein A-I (ApoA-I) is the best lipoprotein discriminator of CHD risk in non-diabetic populations, but the issue has never been investigated in type 2 diabetes.

METHODS

In 2,627 participants without known vascular disease in the Collaborative Atorvastatin Diabetes Study, ApoB, ApoA-I, LDL-cholesterol (LDLC) and HDL-cholesterol (HDLC) were assayed at baseline.

RESULTS

There were 108 CHD and 59 stroke endpoints over 3.9 years. The ApoB:A-I ratio at baseline was the lipoprotein variable most closely predicting CHD risk both by comparison of the hazard ratio for a 1 SD change or tertiles of frequency distribution. The areas under the receiver-operator curve for the ApoB:ApoA-I and the LDLC to HDLC [corrected] ratios, although not significantly different from each other, were greater (p = 0.0005 and p = 0.0125 respectively) than that of non-HDLC:HDLC. The 27% decrease in the ApoB:ApoA-I ratio on atorvastatin predicted a 32% (95% CI 5.4-51.2%) risk reduction in CHD, close to the 36% decrease observed. Neither the ApoB:ApoA-I nor any other lipoprotein concentration or ratio predicted the stroke outcome.

CONCLUSIONS/INTERPRETATION

Overall, the ApoB:ApoA-I ratio improved on the non-HDLC:HDLC ratio in predicting CHD, but, depending on the assessment chosen, its superiority over LDLC:HDLC may be marginal. The statin-induced decrease in stroke risk may not be lipoprotein mediated.

TRIAL REGISTRATION

ClinicalTrials.gov NCT00327418.

FUNDING

The study was supported by unrestricted grants from Diabetes UK, the Department of Health and Pfizer to the University of Manchester and to University College, London.

Medical therapy for acromegaly

Both somatostatin analogues, which bind to the somatostatin receptor subtypes 2 and 5, and dopamine agonists, which are specific for the D2 receptor, have been used to treat acromegaly. Each of these classes of drugs contains several compounds that vary in duration of action, efficacy, and side effect profile. Although somatostatin analogues reduce GH levels and alleviate symptoms in most patients and restore IGF-1 levels to normal in 60% to 65% of patients, tumor shrinkage is limited to 40% of patients. evidence in the literature supports the use of these medications as secondary therapy in patients with acromegaly who have had surgery and who continue to have elevated GH levels (above 2 ng/mL during an oral glucose tolerance test) with or without IGF-1 concentrations that are above the upper limit of normal for age. In addition, medical therapy indicated in patients who refuse surgery and in patients who are poor surgical candidates. The controversial question is whether medical therapy should be an option for primary treatment of the acromegalic patient. Currently, ther are no data from prospective randomized trials comparing the effects of surgery versus somatostatin analogues as first-line therapy for for newly diagnosed acromegalic patients. Limited data from nonrandomized studies demonstrate that somatostatin analogues are effective long-term in suppressing GH and reducing IGF-1 into the normal range in approximately two-thirds of patients who have never undergone previous treatment. It is still the consensus that patients with GH-secreting microadenomas should undergo surgical resection, because the likelihood of complete cure by an experience neurosurgeon is high, at least 70% or greater. Successful surgical treatment has the advantage of completely removing the tumor in contrast to medical therapy, which rarely produces shrinkage greater than 50% despite the fact that IGF-1 and GH levels may be normal. In patients with macroadenomas of a size and location that suggest that the chance of complete resection is 40% or less, primary treatment with a somatostatin analogue should be considered as one option in the initial management of the patient. Another option in such an individual would be surgical debulking followed by medical therapy, because it is theoretically possible that biochemical cure with medical therapy after surgical debulking might be achieved with lower doses. The cost-effectiveness of these approaches has not yet been determined. Once the decision has been made to begin medical therapy, a choice must be made between dopamine agonists and somatostatin analogues. Most evidence suggests that somatostatin analogues are more effective than dopamine agonists and therefore would be the therapy of choice. In select patients, dopamine agonists, particularly the long-acting agonist cabergoline, may be preferred initially if the patient is unwilling to take injections or if the GH elevations are relatively modest (< 10 ng/mL). Biochemical cure should be assessed by measurement of GH (which can be performed 2 hours after an octreotide injection) and IGF-1 concentrations. The goal of treatment include reduction of of GH below 2 ng/mL and reduction of IGF-1 into the normal range. In patients who do not reach these goals, the dose or frequency of injection of the somatostatin analogue or both should be increased. If such measures are unsuccessful, a dopamine agonist may be added to the medical regimen because some studies suggest that combination therapy may be more effective in select cases than octreotide therapy alone. If such measures are still unsuccessful, other options should be considered, including surgery, pituitary radiation, and medical treatment with investigational drugs.

Octreotide as primary therapy for acromegaly

The effects of octreotide (up to 5 yr) as primary treatment in 26 patients with acromegaly were compared with those in 81 patients with acromegaly who received octreotide as secondary or adjunctive therapy after previous surgery and/or pituitary radiation. These patients were part of a multicenter study that took place between 1989-1995. The study was divided into 3 phases beginning with a 1-month placebo-controlled treatment period followed by a 1-month washout period. In the second phase, patients were randomized to treatment with either 100 or 250 micrograms octreotide, sc, every 8 h for 6 months. Octreotide was then discontinued for 1 month and reinitiated at the lower dose for a total mean treatment duration of 39 months. The dose was titrated by each investigator to improve each patient's individual response, which included improvement in symptoms and signs of acromegaly as well as reduction of GH and insulin-like growth factor I (IGF-I) into the normal range. In the second phase of the study, in which patients were randomized to either 100 or 250 micrograms octreotide, three times daily, mean integrated GH and IGF-I concentrations after 3 and 6 months were equivalent in the primary and secondary treatment groups. During long term open label treatment, mean GH fell from 32.7 +/- 5.2 to 6.0 +/- 1.7 micrograms/L 2 h after octreotide injection in the primary therapy group and remained suppressed for a mean period of 24 months (range, 3-60 months). The mean final daily dose was 777 micrograms. In the patients receiving secondary treatment, mean GH fell from 30.2 +/- 7.6 to 5.6 +/- 1.1 micrograms/L after 3 months and remained suppressed for the remainder of the study (average dose, 635 micrograms daily). Mean IGF-I concentrations fell from 5.2 +/- 0.5 x 10(3) U/L (primary treatment group) and 4.7 +/- 0.4 x 10(3) U/L (secondary treatment group) to a mean of 2.2 +/- 0.3 x 10(3) U/L in both groups after 3 months of open label treatment and remained suppressed. IGF-I was reduced into the normal range during at least half of the study visits in 68% of the primary treatment group and in 62% of the secondary treatment group. Patients whose GH levels fell to at least 2 SD below the baseline mean GH were considered responders. There was no significant difference in the percentage of responders in the primary and secondary treatment groups (70% vs. 61%), nor was there a statistical difference in the mean GH concentrations between the groups. Symptoms of headache, increased perspiration, fatigue, and joint pain were reported at baseline by 46%, 73%, 69%, and 85%, respectively, of patients in the primary therapy group and improved during 3 yr of octreotide treatment in 50-100%. Similarly, these acromegaly-related symptoms were reported by 62%, 58%, 78%, and 60% of patients in the secondary therapy group, and improvement was noted in 62-88%. Pituitary magnetic resonance imaging scans were available in 13 of 26 patients in the primary treatment group before and after 6 months of octreotide treatment. Tumor shrinkage was observed in 6 of 13 patients, with reduction in tumor volume greater than 25% in only 3. Of 6 patients with documented tumor shrinkage, IGF-I was reduced into the normal range in 4 patients. Of the 7 remaining patients in whom tumor shrinkage was less than 10%, IGF-I was reduced into the normal range in 4 patients. Of the 7 remaining patients in whom tumor shrinkage was less than 10%, IGF-I was reduced into the normal range in 5 patients. The degree of tumor shrinkage did not correlate with the percent reduction in IGF-I or GH. In summary, octreotide was equally effective in 26 previously untreated acromegalic patients (primary treatment group) and 81 patients previously treated with either surgery or pituitary radiation (secondary treatment group). These observations call into question the current practice of surgical resection of all newly diagnosed GH-secreting pituitary adenomas regardless of the likelihood of cure. (AB

Safety and efficacy of long-term octreotide therapy of acromegaly: results of a multicenter trial in 103 patients--a clinical research center study

One hundred and three acromegalic patients from 14 medical centers were enrolled in this study to determine the efficacy and safety of the somatostatin analog, octreotide acetate, during long term treatment. Seventy percent of the patients had undergone previous surgery or radiation treatment. Octreotide was initiated at a dose of 100 micrograms, sc, every 8 h and gradually increased to a maximum of 1500 micrograms daily depending upon the individual patient's clinical and biochemical response [GH and insulin-like growth factor I (IGF-I) reduction]. The mean duration of treatment was 24 months (range, 3-30 months). However, most patients were treated for a mean of 30 months, because this study took place after an initial 6-month study previously reported. Mean serum GH fell from 30.9 micrograms/L (range, 2.7-350) to 5.7 micrograms/L (range, 0.6-59) at the 3 months visit and remained suppressed (P < 0.001). Plasma IGF-I concentrations were also significantly reduced and remained in the normal range for at least half of the treatment visits in 56 of 87 patients (64%) treated for 12-30 months. Patients with higher initial GH concentrations were less likely to normalize IGF-I concentrations during treatment (P < 0.001). There was no evidence of drug tachyphylaxis in those patients who continued taking stable doses of medication. With some exceptions, dose increments above 800 micrograms daily in 31 patients did not provide additional benefit in terms of GH and IGF-I reduction. Headache, excessive perspiration, fatigue, and joint pain were ameliorated in 83-95% of patients. Mean finger circumference was decreased significantly at the 12 month visit (P < 0.05). The most common adverse events reported were diarrhea, abdominal discomfort, loose stools, and nausea; these symptoms usually disappeared within 3 months of treatment. Five patients discontinued octreotide because of adverse events. Of 102 patients with normal baseline ultrasound examinations of the gallbladder, 24 patients (23.5%) developed gallstones (usually during the first year of treatment), and 21 patients developed sludge alone. Gallstone formation was not related to the dose of octreotide. Most patients with cholelithiasis were asymptomatic, and none developed cholecystitis. These observations suggest that octreotide is a valuable long term medical treatment for acromegaly.

Intact and amino-terminally shortened forms of insulin-like growth factor I induce mammary gland differentiation and development

Growth hormone (GH) plays a role in regulating growth and differentiation of immature glandular structures in the mammary gland, but the mechanisms by which the hormone exerts these effects are unknown. We have previously found that GH stimulates insulin-like growth factor I (IGF-I) I mRNA production within the mammary glands of hypophysectomized rats. In this study we set out to determine if IGF-I administration could mimic the action of GH in initiating mammary gland differentiation and development. Two forms of IGF-I, intact and amino-terminally shortened [des-(1-3)-IGF-I], were found to induce the development of terminal end buds and the formation of alveolar structures in the mammary glands of hypophysectomized, castrated, and estradiol-treated sexually immature male rats. The effect of both forms of IGF-I was similar to that obtained with human GH, but the truncated form was at least 5 times more potent than intact IGF-I. These findings suggest that the inductive effect of GH on glandular differentiation is mediated by the GH-induced production of IGF-I or a related molecule within the mammary gland itself.

A descriptive study of sibling visitation in the NICU

Factors that foster sibling involvement with the newborn and the early stages of the attachment process were cited by key experts: Brazelton and Kennell viewed organization as an essential component of any successful sibling involvement program; Rait and Barnard saw recognition of siblings' needs as vital in any effort to help them process the entire experience; and Consolvo recognized attitudinal issues as playing a key role in implementing and accepting sibling involvement. The interviews with the head nurses identified some common themes within their unique programs that centered around families, the NICU atmosphere, and the promotion of sibling activities. A liberal visiting policy was recognized in all the units, promoting the concept of family-centered care and allowing parents, siblings, and grandparents the opportunity to be supportive together. The flip side of having such a liberal visitation program was the sibling's exposure to a noisy, hurried atmosphere and parents' high anxiety level and the staff's concern about the preterm infant being subjected to a noisier, brighter, more stressful environment and possible exposure to infectious disease. The promotion of sibling activities was a recurring theme in all the head nurse interviews. Examples varied from unit to unit, but the implementation of growth and development principles, along with creativity and innovation, indicated that the overall efforts of these units in promoting sibling involvement was commendable.

Non-lactogenic effects of growth hormone on growth and insulin-like growth factor-I messenger ribonucleic acid of rat mammary gland

In contrast to established dogma that PRL is central in mammary development, and GH mimics PRL in affecting growth because of structural similarities, we found that both hGH, which is lactogenic, and rGH, which is non-lactogenic, were significantly more potent than hPRL and rPRL in stimulating mammary growth in rats. Additionally, hGH was more potent than hPRL in increasing mammary IGF-I mRNA content. These data indicate that GH has separate effects on parameters of mammary gland growth, suggesting an independent role for GH in mammary growth.

Effect of CV 205-502 in hyperprolactinaemic patients intolerant of bromocriptine

CV 205-502 (Sandoz), an octahydrobenzol [g]quinoline, is a long-acting dopamine agonist which inhibits prolactin secretion. We conducted a phase 2 clinical study in 10 hyperprolactinaemic women (nine of whom were previously intolerant of bromocriptine) in order to determine (1) the dose at which CV 205-502 exerted its prolactin-lowering effect; (2) the nature of adverse reactions associated with long-term therapy; and (3) whether patients who were intolerant of bromocriptine could tolerate CV 205-502. At first patients were randomized to take initial doses of either 0.02 or 0.05 mg daily at bedtime. Thereafter these doses of medication were gradually increased either to the point of normalizing serum prolactin (to 0.70 IU/l or 20 ng/ml) or to a maximum dose of 0.14 mg daily. The lower initial dose was ineffective and had to be increased in all patients. The higher initial dose (0.05 mg) normalized prolactin in three of five women within 24 h. During chronic administration of the final dose of CV 205-502 (mean 0.09 mg a day), serum prolactin decreased from a mean level of 9.19 +/- 4.9 (SEM) IU/l to a mean level of 1.55 +/- 0.49 IU/l (n = 10 patients). Prolactin was normalized in five patients. Two patients, one of whom had been previously unresponsive to bromocriptine, and another unresponsive to pergolide with regard to prolactin inhibition, were also unresponsive to CV 205-502. Nausea, the side-effect responsible for these patients' previous intolerance of bromocriptine, occurred in six of 10 patients taking CV 205-502 but was much less disabling and did not cause any of the patients to stop this medication. Only one patient taking CV 205-502 discontinued treatment because of adverse effects (light-headedness).

Evidence for a nonprolactin, non-growth-hormone mammary mitogen in the human pituitary gland

To determine whether the human pituitary contains a previously unidentified, nonprolactin (non-hPRL), non-growth-hormone (non-hGH) factor capable of stimulating mammary development, we tested the effects of whole human pituitary extract (hPE) and pituitary extracts depleted of hPRL and hGH ("stripped hPE") in hypophysectomized, castrated estradiol (E2)-treated male rats and rhesus monkeys. Both whole and stripped hPE significantly stimulated rat mammary development (mean scores = 3.3 and 2.0, respectively, on a scale ranging from 0 to 4) in comparison with controls (mean score = 1.0). Mammary development was not due to minute concentrations of hGH or hPRL remaining in stripped hPE because 30- to 100-fold higher concentrations of hGH (Genentech) and 1000-fold higher concentrations of hPRL were required to stimulate significant mammary development. Non-pituitary extracts of human ovary, muscle, and serum, and bovine serum albumin did not stimulate rat mammary gland growth. Trypsin destroyed the mammary mitogenic activity of whole hPE, indicating that the unidentified factor is likely a protein. Mammary growth and development were also stimulated in hypophysectomized, E2-treated monkeys by stripped hPE (mean histological score = 3.25 vs. 1.35 in control animals). Monkeys receiving stripped hPE had undetectable levels of hGH and hPRL in serum sampled over a 24-hr period. These findings suggest that the human pituitary contains a non-hPRL, non-hGH factor that stimulates mammary growth and may be important in normal mammary growth and development and perhaps in breast cancer.

Dopaminergic regulation of alpha-melanocyte-stimulating hormone and N-acetyl-beta-endorphin secretion in the fetal lamb

alpha MSH is present in high concentrations in the intermediate lobe of the fetal pituitary and has been implicated as a regulator of fetal adrenal steroidogenesis and fetal growth. However, there are few data regarding alpha MSH levels in fetal plasma or the control of fetal alpha MSH secretion. We measured alpha MSH immunoactivity in the plasma of chronically catheterized fetal lambs (gestational age, 116-138 days), newborn lambs, and adult sheep both in the baseline state and after dopamine receptor blockade with metoclopramide. The effect of metoclopramide on the release of another proopiomelanocortin-derived peptide, N-acetyl-beta-endorphin (N-acetyl-beta EP), which is synthesized together with alpha MSH in the intermediate lobe, was also studied. Baseline fetal plasma alpha MSH was significantly greater than maternal alpha MSH [35.6 +/- 2.2 (+/- SEM) vs. 10.0 +/- 1.0 pg/ml]. In eight studies in five fetal lambs, alpha MSH rose to a peak level of 121 +/- 23 pg/ml 15 min after metoclopramide administration to the fetus. Simultaneous maternal alpha MSH levels did not change, suggesting that the alpha MSH in fetal plasma was of fetal pituitary origin. Gel filtration of pooled fetal plasma extracts revealed that the alpha MSH immunoactivity eluted in the same position as the alpha MSH standard. Metoclopramide caused the secretion of nearly equimolar amounts of alpha MSH and N-acetyl-beta EP into fetal plasma. In four fetal lambs, basal N-acetyl-beta EP levels of 156 +/- 34 pg/ml rose to 305 +/- 65 pg/ml 15 min after metoclopramide treatment. Metoclopramide also stimulated plasma alpha MSH in newborn and adult sheep. In six newborn lambs, alpha MSH rose from 45.2 +/- 13 to 211 +/- 38 pg/ml 15 min after metoclopramide treatment, whereas in four adult sheep, a basal alpha MSH level of 11.1 +/- 2.2 pg/ml rose to 20.1 +/- 2.7 pg/ml 15 min after metoclopramide. In addition, metoclopramide stimulated fetal and neonatal PRL secretion, but had no effect on plasma vasopressin concentrations or acid-base and blood gas values. These studies indicate that immunoreactive alpha MSH and N-acetyl-beta EP are secreted into ovine fetal plasma and that the secretion of these peptides in the fetus appears to be under tonic dopamine inhibition, as is the case in the adult sheep and newborn lamb.

Suppression of basal and stress-induced prolactin release and stimulation of luteinizing hormone secretion by alpha-melanocyte-stimulating hormone

alpha-MSH and beta-endorphin, both synthesized from a common precursor, have opposite behavioral actions. In order to determine if these peptides have opposite effects on pituitary function, basal LH secretion and basal and stress-induced prolactin release were studied in adult male rats after intraventricular injection of alpha-MSH. Each rat also received intraventricular saline in order to serve as its own control. 18 micrograms alpha-MSH stimulated plasma LH from 16.5 +/- 2.5 (SEM) ng/ml to a peak of 27.2 +/- 4.0 and 26.0 +/- 4.9 ng/ml at 5 and 10 min, and suppressed prolactin from 3.5 +/- 0.7 ng/ml to 1.3 +/- 0.1 and 1.2 +/- 0.1 ng/ml at 15 and 30 min. Intraventricular alpha-MSH also significantly blunted the prolactin rise associated with the stress of swimming. 10 and 20 min after the onset of swimming, prolactin levels in rats pretreated with alpha-MSH were significantly diminished: 7.4 +/- 1.5 and 6.5 +/- 2.0 ng/ml vs 23.8 +/- 3.6 and 15.2 +/- 2.8 after normal saline. Similarly, des-acetyl alpha-MSH which is the predominant form of alpha-MSH in the hypothalamus, diminished the stress-induced prolactin rise from 18.4 +/- 5.3 and 11.2 +/- 3.4 ng/ml at 10 and 20 min to 10.0 +/- 2.4 and 5.5 +/- 1.6 ng/ml. We conclude that centrally administered alpha-MSH stimulates LH and suppresses basal and stress-induced prolactin release in male rats. These actions are opposite to those previously shown for beta-endorphin and suggest that alpha-MSH may antagonize the effects of beta-endorphin on pituitary function.

Prolactin response to breast stimulation in lactating women is not mediated by endogenous opioids

Several reports have shown that the prolactin response to suckling in rats can be blunted by administration of the opiate antagonist naloxone. In order to investigate whether the prolactin response to breast stimulation in women is similarly affected by naloxone, nine healthy lactating women participated in 10 studies. Each woman served as her own control and was studied on two occasions, receiving pretreatment with either saline solution or naloxone. Prolactin was measured in the baseline state and for 60 minutes after the onset of a 20-minute period of nipple stimulation by use of the Egnell mechanical breast pump. Neither baseline nor stimulated prolactin values were different by paired t test. Thus, in contrast to rats, an opioid pathway does not appear to be involved in the prolactin response to suckling in humans.

Adrenocorticotropin immunoactivity in monkey hypophyseal portal blood

ACTH was measured with both C-terminal and midportion antibodies in monkey hypophyseal portal plasma, and compared to levels in monkey peripheral plasma, medial basal hypothalamus, and anterior pituitary. In nine female monkeys, mean hypophyseal portal blood C-terminal ACTH immunoactivity was 5290 +/- 2010 (SEM) pg/ml, whereas the mean midportion ACTH level was 949 +/- 178 pg/ml. These immunoactivities were not lower in two monkeys that were completely hypophysectomized 30 min before portal blood collection. The ratio of C-terminal to midportion ACTH immunoactivity was 4:1 in two monkey medial basal hypothalami, and 1:1 in four monkey anterior pituitary glands. Gel filtration of hypophyseal portal plasma extract and of medial basal hypothalamus showed that the C-terminal ACTH immunoactivity eluted in the same position as the corticotropin-like-intermediate lobe peptide standard. The similarity of the C-terminal to midportion ACTH ratios in monkey medial basal hypothalamus and portal blood, and the observation that ACTH immunoactivity was not significantly lower in two hypophysectomized monkeys suggests that portal blood C-terminal ACTH immunoactivity is of hypothalamic rather than pituitary origin. We conclude that monkey hypophyseal portal blood contains high levels of a C-terminal fragment of ACTH, which coelutes with corticotropin-like intermediate lobe peptide on gel filtration, and which is secreted from the brain directly into the portal circulation.

Endocrine function in children with intrasellar and suprasellar neoplasms: before and after therapy

The records of 23 children with intrasellar and suprasellar neoplasms were reviewed for the results of endocrine evaluations before and after treatment with surgery, irradiation, and/or chemotherapy. Deficiency of at least one pituitary hormone was present in 13 patients before treatment and in 22 patients after treatment. Growth hormone deficiency, the most common endocrine abnormality in the pretreatment period, was present in 12 of 17 patients before treatment and in 17 of 21 patients after treatment. In the posttreatment period, thyroid-stimulating hormone deficiency was as common as growth hormone deficiency and was found in 17 of 21 patients evaluated. Hypothalamic-pituitary-gonadal dysfunction and deficiencies of adrenocorticotropic hormone and antidiuretic hormone also occurred in both the pretreatment and posttreatment periods, but were less common.

Acetaldehyde adducts with hemoglobin

Clinical studies on the minor hemoglobins (hemoglobin A1a-c) have suggested that a novel adduct may form in people abusing alcohol. Such patients were found to have an elevated concentration of minor hemoglobins, but normal or subnormal amounts of glycosylated hemoglobin (hemoglobin A1c) as determined by radioimmunoassay, Acetaldehyde, a reactive metabolite of ethanol, was postulated to form adducts with hemoglobin A that change its chromatographic properties. At physiological concentrations, acetaldehyde was found to form adducts with hemoglobin that were stable to extensive dialysis for several days. The amount of hemoglobin adducts formed were a function of the concentration and number of exposures to acetaldehyde. The reaction of acetaldehyde with hemoglobin A produced chromatographic variants, some of which migrated in the hemoglobin A1a-c region. Analysis of stable acetaldehyde-hemoglobin adducts demonstrated that valine, lysine, and tyrosine residues of globin were sites of reaction. The acetaldehyde-modified amino acid residues appear to exist in interconvertible conformations, only one of which is reducible by sodium borohydride. The amount of these adducts was found to be significantly elevated in hemoglobin from alcoholics as compared with normal volunteers.