The development and evaluation of a community-based clinical diagnosis tool and treatment regimen for postpartum sepsis in Bangladesh and Pakistan.

BACKGROUND: Postpartum sepsis accounts for most maternal deaths between three and seven days postpartum, when most mothers, even those who deliver in facilities, are at home. Case fatality rates for untreated women are very high. Newborns of ill women have substantially higher infection risk. METHODS/DESIGN: The objectives of this study are to: (1) create, field-test and validate a tool for community health workers to improve diagnostic accuracy of suspected puerperal sepsis; (2) measure incidence and identify associated risk factors and; (3) describe etiologic agents responsible and antibacterial susceptibility patterns. This prospective cohort study builds on the Aetiology of Neonatal Infection in South Asia study in three sites: Sylhet, Bangladesh and Karachi and Matiari, Pakistan. Formative research determined local knowledge of symptoms and signs of postpartum sepsis, and a systematic literature review was conducted to design a diagnostic tool for community health workers to use during ten postpartum home visits. Suspected postpartum sepsis cases were referred to study physicians for independent assessment, which permitted validation of the tool. Clinical specimens, including urine, blood, and endometrial material, were collected for etiologic assessment and antibiotic sensitivity. All women with puerperal sepsis were given appropriate antibiotics. DISCUSSION: This is the first large population-based study to expand community-based surveillance for diagnoses, referral and treatment of newborn sepsis to include maternal postpartum sepsis. Study activities will lead to development and validation of a diagnostic tool for use by community health workers in resource-poor countries. Understanding the epidemiology and microbiology of postpartum sepsis will inform prevention and treatment strategies and improve understanding of linkages between maternal and neonatal infections.

Characterization and anti-settlement aspects of surface micro-structures from Cancer pagurus.

Tuning surface and material properties to inhibit or prevent settlement and attachment of microorganisms is of interest for applications such as antifouling technologies. Here, optimization of nano- and microscale structures on immersed surfaces can be utilized to improve cell removal while reducing adhesion strength and the likelihood of initial cellular attachment. Engineered surfaces capable of controlling cellular behaviour under natural conditions are challenging to design due to the diversity of attaching cell types in environments such as marine waters, where many variations in cell shape, size and adhesion strategy exist. Nevertheless, understanding interactions between a cell and a potential substrate for adhesion, including topographically driven settlement cues, offers a route to designing surfaces capable of controlling cell settlement. Biomimetic design of artificial surfaces, based upon microscale features from natural surfaces, can be utilized as model surfaces to understand cell-surface interactions. The microscale surface features of the carapace from the crustacean Cancer pagurus has been previously found to influence the rate of attachment of particular organisms when compared to smooth controls. However, the nature of microscale topographic features from C. pagurus have not been examined in sufficient detail to allow design of biomimetic surfaces. In this work, the spatial distribution, chemical composition, size and shape descriptors of microscale surface features from C. pagurus are characterized in detail for the first time. Additionally, the influence of topography from C. pagurus on the settlement of marine diatoms is examined under field conditions.

A finite-element model predicts thermal damage in cutaneous contact burns.

Thermal injury results from exposure of skin elements to an externally applied heat source. Finite-element analysis of heat transfer in cutaneous burns allows for an accurate prediction of tissue time-temperature relationships throughout the exposed tissue. A two-dimensional, axisymmetric, finite-element model of a contact burn was constructed, and damage integrals were calculated by applying the Arrhenius equation to the time-temperature profiles at each point. The epidermis, dermis, and subcutaneous fat were modeled as uniform elements with distinct thermal properties. Heated aluminum blocks were applied to Yorkshire pigs for 10 to 80 seconds to produce contact burns. Wound biopsies taken at 1, 24, and 48 hours were examined histologically and measured for the depth of burn. A significant deepening of the gelatinized tissue was observed in tissue taken from 1 hour to 24 hours. The finite-element prediction of cutaneous contact burn damage correlated well with histologic observations in this porcine model.

Cultured human sole-derived keratinocyte grafts re-express site-specific differentiation after transplantation.

Cultured epithelial autografts (CEA) derived from sole skin were transplanted to full-thickness wounds excised to muscle fascia over a variety of diverse body sites in 12 pediatric patients treated for acute burns or giant congenital nevi. The skin regenerated from the grafts was biopsied from 7 days to 6 years after grafting. The resultant epidermal phenotype was analyzed histologically and by immunohistochemical localization of keratin 9 (K9) as objective evidence of sole-type site-specific differentiation. Expression of K9 was also verified by one-dimensional gel electrophoresis of epidermal cytoskeletal extracts and K9 immunoblot analysis. Grafts prepared from epidermis of axilla; groin or foreskin and transplanted to wounds of comparable depth in an identical manner in the same patients served as controls of postgrafting differentiation. Biopsies of sole skin from amputation specimens from patients of comparable age served as normal positive controls, and biopsies of nonsole skin from patients of comparable age served as normal negative controls. As early as 2 weeks postgrafting, the histologic appearance of sole-derived CEA differed substantively from that of axilla- or groin-derived CEA controls and displayed a phenotype characteristic of sole skin with a thick compact stratum corneum, a thick stratum granulosum, and a distinct stratum lucidum. In sole-derived grafts rete ridges regenerated within 2 months postgrafting, whereas nonsole-derived grafts required 4-6 months for rete ridge regeneration. Once acquired, the sole skin phenotype was maintained long-term by all sole-derived CEA. In vitro, sole-derived keratinocytes synthesized little, if any, K9. However, within 7 days after grafting, K9 synthesis by multiple suprabasal keratinocytes was seen within the epidermis regenerated from sole-derived CEA. Protein of K9 appeared progressively more diffuse throughout the suprabasal layers, attaining a confluent pattern of expression comparable to normal controls of sole skin by 6 to 12 months postgrafting, and the confluent pattern of suprabasal K9 synthesis was maintained long-term. The results demonstrate that site-specific differentiation is an intrinsic property of postnatal human keratinocytes and can be expressed and maintained in a permissive environment in the absence of dermal tissue.

Engineering a skin replacement.

Major skin loss from trauma or burns cannot always be replaced with the patient's own skin. An engineered skin replacement would restore the barrier function of the skin, remain permanently on the wound, and minimize late functional complications of wound contraction. Cultured epithelial autograft (CEA) sheets reproduce the epidermis' function and have been used in burn patients to close large wounds. There are several promising avenues for dermal replacement, but none has yet had wide clinical application.

Skin substitutes in upper extremity burns.

Four techniques for permanent skin replacement with skin substitutes are described. A claim of superiority to conventional skin grafting on upper extremity and hand burns is not made, but some clinical observations and histologic evidence of different healing characteristics are shown. The composite grafts described appear to effectively replace the bilayered structure of skin and seem to have good subjective resistance to shear forces. If the skin replacement is durable and heals with less scarring than conventional skin grafts, the inexorable course to stiffness and contracture may be altered. Further basic science and clinical investigation may provide us with a better way of managing these difficult problems.

Cultured epithelial autografts for giant congenital nevi.

Eight pediatric patients with giant congenital nevi confluent over 21 to 51 percent body surface area were treated by excision and grafting. The nevus was excised to the muscle fascia, and the open wound was grafted with cultured epithelial autografts and split-thickness skin grafts. The patients have been followed from 17 to 56 months. Seventeen operations were performed in the eight patients, excising a mean of 6.9 percent body surface area at each procedure. The mean duration of anesthesia was 3.7 hours, and the mean operative blood loss was 12.3 percent estimated blood volume. The mean "take" for the cultured epithelial autografts was 68 percent, and for the split-thickness skin grafts, 84 percent. Epithelialization of open wound areas adjacent to the grafts was somewhat slower for the cultured epithelial autografts than for the split-thickness skin grafts, but it led to a healed wound in all patients except one. Ten of the 17 areas grafted with cultured epithelial autografts resulted in small open wounds that required regrafting. Wound contraction under the cultured epithelial autografts and under split-thickness skin grafts was similar and depended more on the anatomic site grafted than on the type of graft employed. in 16 of 17 operations, the cultured epithelium remained as a permanent, durable skin coverage. The use of cultured epithelial autografts allowed a larger area of excision than would have been possible with split-thickness skin grafts alone and, therefore, a more rapid removal of nevus. Cultured epithelial autograft are an important new technique in the care of patients with giant congenital nevi.

Skin regenerated from cultured epithelial autografts on full-thickness burn wounds from 6 days to 5 years after grafting. A light, electron microscopic and immunohistochemical study.

Regeneration of skin from cultured keratinocyte autografts used in the treatment of full-thickness burn wounds was studied in 21 pediatric patients from 6 days to 5 years after grafting. Findings were compared both to controls of age- and site-matched normal skin and to controls for epithelial wound-healing, re-epithelialized interstices of meshed split-thickness skin grafts of comparable postgrafting age. Six days after transplantation, a mildly hypertrophic, flat epidermis with all normal strata had regenerated, and the process of de novo dermal-epidermal junction formation had begun. Hemidesmosomes, basal lamina, and anchoring fibrils reformed conjointly in punctate fashion along the attachment face of the grafts. Within 3 to 4 weeks, the dermal-epidermal junction was complete, but full maturation of anchoring fibrils required more than a year. The process was comparable to that observed in meshed graft interstices. Rete ridges regenerated from 6 weeks to 1 year after grafting. The subjacent connective tissue initially healed to form normal scar, but it remodeled dramatically, regenerated elastin, and resembled a true dermis within 4 to 5 years. Meshed-graft interstice controls showed no rete ridge regeneration, subepithelial connective tissue remodeling, or elastin production up to 5 years after grafting. Langerhans cells repopulated grafts within 1 week, and normal population densities were reached within 2 to 6 months. After 1 year, Langerhans cell densities were increased compared with normal skin but were lower than those in age-matched meshed graft controls. Melanocytes were present in cultures at the time of transplantation, but functional epidermal melanin units were not seen in groin- or axilla-derived grafts for 6 to 8 weeks or in sole-derived epidermis until a year or more after transplantation. Normal histologic features were maintained for years after grafting. Transitory pathologic changes including parakeratosis, dyskeratosis, and intraepithelial friction blister formation were infrequently observed. No dysplastic or premalignant changes were seen.

Cultured epithelium as a skin substitute.

The history of techniques for the replacement of lost skin is reviewed, including the current research in the use of synthetic dermal substitutes, skin allografts and immunosuppression, and tissue-cultured epithelial autografts. Developments in each of these three areas are encouraging steps toward the development of a skin substitute that would be immediately available for coverage of even massive areas of skin loss.

Anergy, immunosuppressive serum, and impaired lymphocyte blastogenesis in burn patients.

Skin testing with four recall antigens was performed serially in 21 patients after a major thermal burn. We looked for a correlation between the occurrence of anergy, the presence of immunosuppressive serum, and the impairment of the lymphocyte-proliferative response to phytohemagglutinin (PHA). Serum cortisol, endotoxin, and prostaglandin E2 (PGE2) levels were also measured in the serum or plasma. When anergy developed, it became apparent early in the course of the illness. It did not correlate closely with the severity of the burn, but was associated with mortality. There was a good correlation between anergy and coexisting serum suppression of lymphocyte activation in vitro. This serum immunosuppressive activity was not related to serum cortisol, PGE2, or plasma endotoxin levels. Anergy also correlated with coexistent impairment of patient peripheral blood lymphocyte activation by PHA. These results suggest that both immunosuppressive serum and an impaired lymphocyte response to mitogens are associated with anergy in burn patients and confirm that the development of anergy is an index of poor prognosis.

Suppressive serum, suppressor lymphocytes, and death from burns.

Both suppressor lymphocytes and serum immunosuppressive factors have been found in patients who have had major thermal burns, and may inhibit host resistance to the bacteria invariably present in burn wounds. However, the relationship and clinical importance of these two manifestations of impaired immune reactivity are poorly understood. Eighteen patients (aged 20-84 years) with full thickness burns of varying severity have been studied, and the clinical course related to the presence of nonspecific immunosuppressive serum and circulating suppressor lymphocytes. Serum factors capable of suppressing the phytohemagglutinin (PHA) response of normal lymphocytes usually appeared early and were detected in 15 of the 18 patients at some time during the illness. Thirteen of these patients developed systemic infection. Depression of the PHA response of peripheral blood lymphocytes was much less common and was associated with this finding died. No patients who did not have severe depression of the lymphocyte response to PHA died. Nonadherent leukocyte (NA leukocyte) populations exhibiting a depressed PHA response were capable of suppressing the PHA response of normal human lymphocytes and, therefore, contained suppressor cells.

The effects of glucagon on protein metabolism in normal man.

Plasma glucagon rises after major injury and could act to increase gluconeogenesis and ureagenesis in the post-traumatic state. This study documents the effect of prolonged glucagon infusion on ureagenesis and nitrogen excretion, as well as possible sources of the increased ureagenesis, in normal man. Four healthy men fasted for 6 days during intravenous infusion of glucose (750 gmday), establishing a steady state of minimal ureagenesis. Glucagon (1 mg/day) then was added to the infusion for 5 days. Glucose alone was given for the final 2 days. Forearm muscle flux of metabolites was determined by standard arterial-deep venous sampling and capacitance plethysmography. Glucagon concentration was suppressed during glucose infusion (11 +/- 13 pg/ml) and rose to levels seen in subjects with major trauma during glucagon infusion (669 +/- 138 pg/ml). Glucose infusion stabilized urine nitrogen excretion at 1.54 +/- 0.42 gm of N/sq m/day. Nitrogen excretion increased to 2.40 +/- 0.53 gm of N/sq m/day with glucagon infusion, with urea accounting for the increased excretion. Excretion of 3-methylhistidine was unchanged. Plasma amino acid concentration was strikingly reduced on the first day of glucagon infusion, where it stabilized. Forearm flux showed a slight net release of amino acid nitrogen during glucose infusion. Addition of glucagon to the glucose infusion resulted in a net uptake of nitrogen by forearm skeletal muscle. These evidences strong suggest that glucagon infusion in normal man increases ureagenesis, not only at the expense of the free amino acid pool, but by the hydrolysis of visceral protein as well, with muscle protein being maintained.

Endocarditis with the indwelling balloon-tipped pulmonary artery catheter in burn patients.

The postmortem finding of acute right-sided bacterial endocarditis in a burn patient monitored with an indwelling pulmonary artery (Swan-Ganz) catheter for 14 days prompted a review of burn autopsies in which the catheter had been used. Autopsies of six consecutive burn patients monitored with a pulmonary artery catheter and who then died showed septic or aseptic endocarditis. In two of the six patients, right-sided staphylococcal endocarditis was the anatomic cause of death. In the remaining four, the lesions were aseptic thrombotic vegetations involving primarily the right atrium, tricuspid valve, right ventricle, and pulmonic valve. Several factors in the severely burned patient would favor endocarditis where a foreign object impacts on the heart valves. These include intermittent bacteremia, hypercoagulability, hyperdynamic cardiovascular function, and the use of antibiotics resulting in resistant strains. While an indwelling pulmonary artery catheter can provide useful monitoring information, it is sometimes responsible for serious complications in burned or septic patients.

The influence of clinically undetectable pulmonary edema on small airway closure in the dog.

The relationship between elevated pulmonary extravascular water volume(PEWV)and small airway closure was examined. The slow accumulation of lung water was achieved by a combination of pulmonary venous hypertension and mild hemodilution. PEWV was measured using a double indicator method based on the differential right to left transit time for simultaneously injected Evans blue dye and tritiated water. Trapped gas volume (VTG) was measured by the helium equilibration technique. Clinically undetectable levels of pulmonary engorgement and edema were reproducibly associated with an increase in gas trapping. Positive end expiratory pressure reduced, but did not abolish, edema formation. Evaluation of airway closure, with consequent gas trapping and pulmonary shunting, is currently non-invasive, simple and safe. Determination of gas trapping or closing volume should be incorporated into the rountine pre-operative evaluation of patients prior to major surgery.

Crystalloid versus colloid hemodilution in man.

The effects of intraoperative prebleeding and hemodilution with lactated Ringer's solution on hemodynamics, oxygen transport and lung water were studied in four patients undergoing extensive surgical procedures. The results were contrasted with those previously obtained from hemodilution of a group of patients with Plasmanate. The mean volumes bled were 1,950 milliliters in the lactated Ringer's solution group and 1,697 milliliters in the Plasmanate group. The posthemodilution hematocrit values were 26 per cent and 25 per cent, respectively. Cardiac output increased to only 115 per cent of the base line in the lactated Ringer's solution group but to 161 per cent in the Plasmanate group. Systemic oxygen transport was reduced to 80.0 per cent in the lactated Ringer's solution group, while in the Plasmanate group, it rose to 109.7 per cent. The lung water and alveolar arterial oxygen gradients were increased to 134 and 112 per cent in the lactated Ringer's solution group, whereas in the Plasmanate group, the lung water was reduced to 82 per cent, and the arterial alveolar oxygen gradient was reduced to 75 per cent. In these selected patients, hemodilution was well tolerated. Plasmanate hemodilution resulted in better compensatory changes than did hemodilution with lactated Ringer's solution in terms of oxygen transport and changes in lung water.

Rapid estimation of pulmonary extravascular water with an instream catheter.

An arterial catheter-bearing external conductivity electrodes and a thermistor was used for measurement of lung thermal volume (LTV) by the double-indicator method. Ten milliliters of 3% saline at room temperature were injected, dilution curves measured, and LTV calculated as mean transit time difference, less thermistor time constant, times cardiac output (CO). Comparisons were made, in dogs, between LTV, pulmonary extravascular lung water with Evans blue and tritiated water (PEVWtho), and weighed lung water (WLW). Pulmonary edema was induced with dextran and epinephrine. CO was measured by thermodilution in both the pulmonary artery (PA) and aorta (AO) and dye dilution in the AO. CO from dye dilution was compared with thermodilution (aortic detection) to detect irreversible loss of thermal indicator. Comparisons showed good correspondence of dye and thermal curves (Y = 0.91X - 0.16 1/min; r = 0.93). LTV is about 120% of WLW in near normal lungs, 90% of WLW in extreme edema. PEVWtho was 60-70% WLW.

Intraoperative prebleeding in man: effect of colloid hemodilution on blood volume, lung water, hemodynamics, and oxygen transport.

The effects of acute normovolemic hemodilution on lung water, blood volume, hemodynamics, and oxygen transport were studied. The subjects were six patients undergoing major operations, with prebleeding and hemodilution under fluoroxene and nitrous oxide anesthesia. The menatocrit was reduced form 43 to 25 percent in one step, with simultaneous infusion of Plamanate and lactated Ringer's solution. Blood volume was expanded by 5 percent by the hemodilution. The major compensation was a striking rise in cardiac output to 161 percent. Systemic oxygen transport (CO times arterial O2 content) increased despite the marked fall in oxygen-carrying capacity, and the arteriovenous O2 content difference decreased. Lung water the aveolararterial (A-a) oxygen differences were reduced. The procedure was well tolerated by this group of selected patients and homologous blood utilization was reduced.

Nitrogen sparing by carbohydrate in man: Intermittent or continuous enteral compared with continuous parenteral glucose.

To test whether oral carbohydrate would provide greater conservation of body protein than would intravenous carbohydrate, healthy normal human subjects were infused with high doses of glucose either continuously intravenously or by nasogastric tube in both continuous and intermittent regimes. Metabolic responses to high calorie, nitrogen-free infusions in normal man were documented in the blood hormone and substrate changes, and protein sparing was assessed by urinary nitrogen excretion. Continuous glucose produced a lower urinary "nitrogen floor" than did the intermittent regime, and intravenous glucose was more effective than was oral glucose. The insulin responses to continuous nasogastric and continuous intravenous glucose were similar, and nitrogen excretion did not differ between those two groups. The increased insulin levels seen with intermittent glucose were not accompanied by greater protein sparing.