Secondary Logo

Journal Logo


Serum Zinc as a Prognostic Tool for Wound Healing in Hip Hemiarthroplasty

Zorrilla, Pedro MD*; Salido, José A MD*; López-Alonso, Antonio MD**; Silva, Agustín MD

Author Information
Clinical Orthopaedics and Related Research: March 2004 - Volume 420 - Issue - p 304-308
  • Free


Malnutrition is common among patients with hip fractures,26,31 and the most immediate postsurgical consequence of this condition is delayed wound healing.9,16 This in turn leads to increased healthcare costs3,12 because of extended hospital stays10 and to a higher incidence of prosthetic infection.6,28,29

The biochemical14,33 (serum albumin < 3.5 g/dL and serum transferrin < 200 mg/dL) and immunologic4,5 (lymphocyte count < 1500 cells/mm3) nutritional parameters that define the nutritional status of patients having surgery are used as factors for predicting delayed healing.19,23,34

Zinc is an essential nutritional element that directly influences wound healing.1,31 Although few studies contemplate serum zinc concentrations,2,15,27 its deficit has been estimated as being high, ranging between 34% and 52%. It forms part of more enzymes and macromolecules than any other isolated nutrient, and its deficiency effects basic cell division, differentiation, and growth processes. In addition, it plays a role in the modulation of keratinocyte integrins32 and in the activation of lysyloxidase,11 which is responsible for the formation of the links between collagen fibers.

The active fraction, exchangeable protein zinc, is found in plasma, where normal levels range from 95 to 130 μg/dL.22 Although certain controversy exists, serum zinc is the most suitable indicator for detecting its deficiency, because other indirect methods have not been proven effective.35

We analyzed the relationship between preoperative serum zinc levels and other biochemical and immunologic nutritional parameters, and delayed wound healing in patients who had hemiarthroplasty for subcapital hip fracture to establish nutritional risk criteria and to determine which patients will experience a delay in healing.


A prospective study was done between January 1, 1998 and September 28, 1999 of patients who had hemiarthroplasty for hip fracture. The patients were included in the order in which they had surgery after admission to the hospital. Ninety-eight procedures were done on 97 patients. The second procedure of the only patient who had surgery on two nonsimultaneous occasions was excluded. The study group was comprised of 97 procedures in 97 patients.

Preoperative serum zinc levels (μg/dL) were determined in all patients by means of atomic absorption spectrophotometry. Preoperative nutritional status was evaluated by determining the levels of serum albumin (g/dL), serum transferrin (mg/dL), and total lymphocyte count (cells/mm3). Demographic variables (age, gender), duration of surgery, and concomitant diseases (diabetes, kidney failure, prior medications) also were analyzed.

For prophylaxis, Cefamandol (Lilly, Madrid, Spain) was given preoperatively and postoperatively. For patients with an allergy to penicillin, vancomycin was used. Calcium nadroparin (Sanofi-Winthrop, Paris, France) was administered for thromboembolic prophylaxis.

In each patient, the Watson-Jones anterolateral approach was used, and incision closure was done with Dexon-0 sutures (Braun-Dexon, Gosport, United Kingdom) for the fascia, Dexon-00 sutures for subcutaneous tissue, and staples for the skin. The staples were removed after 22 days (range, 16–26 days). Removal of drains was done in all patients 48 hours after surgery without any complications. Active joint mobilization was started on the day after surgery.

Direct assessment of wound healing requires manipulation of the same (biopsy specimen to measure hydroxyproline, resistance to traction). As an indirect method, in the current study, delayed healing was defined according to the criteria of Gherini et al13: (1) persistence of serous drainage through the surgical wound beyond postoperative Day 3, and (2) separation of the upper edges of the wound greater than 1 cm transversely or 2 cm longitudinally.

All surgical wounds were evaluated by the same observer, who had no knowledge of the patients’ serum zinc levels.

Wounds were considered to be infected when bacteriologic cultures, which were obtained using standardized methods21 whenever serous drainage appeared, were positive in two consecutive samplings. After removing the dressing and without cleaning the wound, a sample was taken for aerobic culture with a sterile turunda in the designated drainage area, in either the deep or active drainage areas of the wound. Only one pass was made with each sterile swab, which then was placed in the sterile container. Immediately thereafter, a sample was taken for anaerobic culture, using a special culture tube. The wound then was cleaned with an antiseptic and covered with a sterile dressing. The sample was sent to the laboratory immediately.


The statistical method consisted of a univariant analysis of all the independent variables, establishing their relationship with impaired surgical wound healing. The variables that showed a statistically significant relationship were included in a multivariate logistical regression analysis with Version 9 of the SPSS statistical computing (Hispanoportuguesa SPSS, Madrid, Spain). In this type of multivariate analysis, the inclusion of at least 10 patients for each parameter to be estimated is considered sufficient. This would be 10 (k + 1) patients, k being the number of independent variables (including the control subjects) included in the model. Another necessary condition is to have at least 10 patients for each category of the dependent variable.18 In the current study, it was expected to have 30 patients with delayed healing to have no difficulties in applying the univariant parametric tests.

Pearson’s linear correlation test was used for analyzing the association between two quantitative variables.


In 85 patients, a Thompson cemented prosthesis was implanted, whereas in 12 patients a bipolar prosthesis with an uncemented stem (IQL, Valencia, Spain) was used. The mean age of the patients was 80.2 years (range, 62–101 years).

The mean preoperative serum zinc level was 101.51 μg/dL, with a standard deviation of 19.46 μg/dL and a range of 62 to 172 μg/dL. In 41 patients (42.3%), the level was lower than 95 μg/dL. Thirty-three patients (34%) had a preoperative albumin level less than 3.5 g/dL. In 58 patients (59.8%), the lymphocyte count was less than 1500 cells/mm3, and in 38 patients (39.2%), serum transferrin was less than 200 mg/dL.

Serum zinc did not show significant correlation with any of the nutritional parameters studied (p > 0.05 and r < 0.4) according to Pearson’s linear correlation test.

Thirty patients (30.9%) had delayed healing of the surgical wound. This was manifested in all patients by persistence of serous drainage through the wound beyond postoperative Day 3. In two patients, surgical wound dehiscence appeared, which was accompanied by persistent serous drainage. In three patients (3.1%), superficial wound infection was diagnosed, with positive Staphylococcus aureus cultures in two patients and Staphylococcus epidermidis in one patient. In the three patients with persistently positive cultures, local signs of inflammation were observed in the form of erythema and edema along the edges of the wound (not including signs of pain and local heat), signs that are not quantifiable but were always evaluated by the same observer. None of these patients needed surgical cleaning of the wound.

The average postsurgical hospital stay of patients with impaired healing was 7.9 days (range, 6–13 days). The average postsurgical stay of patients who did not have impaired healing was 6.3 days (range, 6–9 days). The differences between both groups were statistically significant when the Student’s t test was applied for the comparison of independent means (p < 0.001).

A univariant analysis was done including all the independent variables, which established their relationship with impaired healing of the wound. Statistical significance was found for serum zinc (p < 0.001) and serum albumin (p = 0.03) (Table 1). No correlations were found among delayed healing and serum transferrin level, lymphocyte count, patient age, gender, duration of surgery, patients with kidney failure or diabetes, and patients who had previous corticosteroid therapy.

Table 1:
Univariate Analysis: T Test

In the multivariate logistic regression analysis, only variables that were proven significant in the univariant study were included as independent variables, namely serum zinc and serum albumin. This analysis revealed an odds ratio of 1.13 for zinc (p < 0.001). Albumin did not show a significant correlation (Table 2). This means that for every microgram per deciliter of decrease in preoperative zinc serum, the probability of a patient having delayed healing increased 1.13 times.

Table 2:
Multivariate Analysis: Logistic Regression

The logistic regression equation for delayed healing probability (P) was:

All the procedures were classified using the aforementioned equation. Table 3 shows a few examples. The prediction had an accuracy of 82.47% in the entire sample, correctly classifying 80% of the patients who experienced delay of healing and 83.58% of the patients who did not experience delay of healing. A prediction greater than 75% is considered good.

Table 3:
Probability of Delayed Healing According Serum Zinc Levels

The sample was categorized into two groups, according to preoperative zinc values: Group 1, zinc ≥ 95 μg/dL; and Group 2, zinc < 95 μg/dL. Group 1 consisted of 56 patients, and 10.7% (six patients) had impaired healing. Group 2 consisted of 41 patients, and 58.5% (24 patients) had delayed healing. The risk of delay was 11.76 times higher (95% confidence interval, 4.11–33.63) in patients in Group 2 than in Group 1.

The test that emerged from dichotomizing the only variable that was negative, serum zinc, had a sensitivity of 80% and specificity of 74%, with negative and positive predictive values of 89% and 58%, respectively (Table 4).

Table 4:


The current study confirms the high incidence of malnutrition among patients with subcapital hip fractures. In a similar study, Koval et al20 found only 18% malnutrition when they used serum albumin as a reference, but this changed to 57% when the lymphocyte count was considered. The 42.3% of patients in the sample had serum zinc levels less than 95 μg/dL, but as in the study of Cruse et al,7 it has not been possible to confirm the assertion25 that low serum zinc level lies in the context of generalized protein-energy undernutrition. Taking into consideration the hospitalized population as a whole, Albina2 reported zinc deficiency at 34%. The results of the current study do not allow one to refer to an everyday zinc deficiency in the patients because fractures bring about a decrease in plasma zinc concentrations30 and an increase in urine elimination. Nevertheless, these results enable discovery of preoperative subclinical deficiencies that may be manifested after surgery.

The current study also confirms that delayed healing is possibly the most common postsurgical morbidity in patients having hemiarthroplasty for subcapital hip fracture. Its incidence, 30.9%, is similar to that observed by Hadden and McFarlane17 (33%). In a more recent study, the incidence observed by Marín et al24 was 23% in a group of patients in whom total hip or knee prostheses had been implanted. The incidence in the current study is somewhat higher, because of strict application of the delay criteria and, because hip fractures are seen in an increasingly aged population with frequent malnutrition.

Impaired healing can be considered a nonserious complication that does not put the patient’s life at risk, but what is much more serious is prosthetic infection. Saleh et al28 confirmed the strong correlation between persistent serous drainage through the surgical wound and the development of a superficial infection and later deep infection. Herein lies the importance of recognizing the population at risk.

Serum zinc (p < 0.001) and albumin (p = 0.008) levels analyzed individually showed significant differences between the groups of patients who had delayed healing and patients who did not have delayed healing. Dickhaut et al8 obtained a sensitivity of 75% by means of the joint determination of albumin and lymphocyte count. For Greene et al,16 an albumin level less than 3.5 g/dL increased the risk of delay sevenfold. Lau et al23 and Vaxman et al34 stated that these parameters provide clinical information on the patient’s nutritional status, but their efficacy in predicting healing complications can be improved by using more sensitive parameters such as serum zinc. In the multivariate analysis, only serum zinc (p < 0.001) showed significant predictive value for delay healing, increasing risk 1.13 times per unit of serum zinc decrease. Several authors refer to zinc as the key nutritional factor in the healing process, but most of these opinions are based on studies on animals.1,22 Plasma zinc level has created interest because of its influence on healing in certain specific disorders, such as in patients with extensive deep burns.15 Cruse et al7 found a close relationship (p < 0.001) between low serum zinc levels and the appearance and delayed healing of bed sores.

The main contribution of this study is that by determining serum zinc in patients with subcapital hip fractures who are scheduled to have hemiarthroplasty, the probability of their experiencing delayed wound healing can be predicted with a sensitivity of 80%. In addition, it may be considered useful for the surgeon to know that, in this disorder, risk of delayed wound healing increased 11.76 times in patients with serum zinc levels less than 95 μg/dL. Consequentially, this could serve to establish presurgical nutritional therapies, which would be useful in patients scheduled for elective surgery. In the case of patients with fractures, who should have immediate intervention, the possibility of supplements being effective in such a short period might be questionable.


We thank the Research Unit of the Complejo Hospitalario Ciudad Real for revision and improvement of the manuscript.


1. Agren MS, Franzen L, Chvapil M. Influence of zinc deficiency on breaking strength of 3-week-old skin incisions in the rat. Acta Chir Scand. 1990;156:667–670.
2. Albina JE. Nutrition and wound healing. J Parenteral Enteral Nutr. 1994;18:367–376.
3. Bernstein LH, Shaw-Stiffel TA, Schorow M, Brouillette R. Financial implications of malnutrition. Clin Lab Med. 1993;13:491–507.
4. Braga M, Baccari P, Scaccabarozzi S, et al. Prognostic role of preoperative nutritional and immunological assessment in the surgical patient. J Parenteral Enteral Nutr. 1988;12:138–142.
5. Christou NV, Tellado-Rodriguez J, Chartrand L, et al. Estimating mortality risk in preoperative patients using immunologic, nutritional, and acute-phase response variables. Ann Surg. 1989;210:69–77.
6. Clarke MT, Longstaff L, Edwards D, Rushton N. Tourniquet-induced wound hypoxia after total knee replacement. J Bone Joint Surg. 2001;83B:40–44.
7. Cruse JM, Lewis RE, Roe DL, et al. Facilitation of immune function, healing of pressure ulcers, and nutritional status in spinal cord injury patients. Exp Mol Pathol. 2000;68:38–54.
8. Dickhaut SC, De Lee JC, Page CP. Nutritional status: Importance in predicting wound healing in amputations. J Bone Joint Surg. 1984;66A:71–75.
9. Ecker ML, Lotke PA. Postoperative care of the total knee patient. Orthop Clin North Am. 1989;20:55–62.
10. Ehrlichman RJ, Seckel BR, Bryan DJ, Moschella CJ. Common complications of wound healing: Prevention and management. Surg Clin North Am. 1991;71:1323–1351.
11. Erdogan M, Ilhan YS, Akkus MA, et al. Effects of l-thyroxine and zinc therapy on wound healing in hypothyroid rats. Acta Chir Belg. 1999;99:72–77.
12. Gariballa SE, Parker SG, Taub N, Castleden CM. Influence of nutritional status on clinical outcome after acute stroke. Am J Clin Nutr. 1998;68:275–281.
13. Gherini S, Vaughn BK, Lombardi AV, Mallory TH. Delayed wound healing and nutritional deficiencies after total hip arthroplasty. Clin Orthop. 1993;293:188–195.
14. Gibbs J, Cull W, Henderson W, et al. Preoperative serum albumin level as a predictor of operative mortality and morbidity: Results from the National VA Surgical Risk Study. Arch Surg. 1999;134:36–42.
15. Gosling P, Rothe HM, Sheehan TM, Hubbard LD. Serum copper and zinc concentrations in patients with burns in relation to burn surface area. J Burn Care Rehabil. 1995;16:481–486.
16. Greene KA, Wilde AH, Stulberg BN. Preoperative nutritional status of total joint patients: Relationship to postoperative wound complications. J Arthroplasty. 1991;4:321–325.
17. Hadden WA, McFarlane AG. A comparative study of closed-wound suction drainage vs. no drainage in total hip arthroplasty. J Arthroplasty. 1990;5S:21–24.
18. Hair JF Jr, Anderson RE, Tatham RL, Black WC. Multiple Regression Analysis. Multivariate Data Analysis With Readings. 4th ed. Upper Saddle River, NJ: Prentice-Hall; 1995:103–110.
19. Kay SP, Moreland JR, Schmitter E. Nutritional status and wound healing in lower extremity amputations. Clin Orthop. 1987;217:253–256.
20. Koval KJ, Maurer SG, Su ET, Aharonoff GB, Zuckerman JD. The effects of nutritional status on outcome after hip fracture. J Orthop Trauma. 1999;13:164–169.
21. Kozier B, Erb G, Blais K, Johnson JY, Temple JS. Cuidado de las Heridas. Técnicas En Enfermería Clínica. Vol 2. 1st ed. Aravaca: McGraw-Hill/Interamericana; 1999:972–974.
22. Lansdown ABG. Zinc in the healing wound. Lancet. 1996;347:706–707.
23. Lau HC, Granick MS, Aisner AM, Solomon MP. Wound care in the elderly patient. Surg Clin North Am. 1994;74:441–463.
24. Marín LA, Salido JA, López A, Silva A. Preoperative nutritional evaluation as a prognostic tool for wound healing. Acta Orthop Scand. 2002;73:2–5.
25. Mora RJ. Malnutrition: Organic and functional consequences. World J Surg. 1999;23:530–535.
26. Patterson BM, Cornell CN, Carbone B, Levine B. Protein depletion and metabolic stress in elderly patients who have a fracture of the hip. J Bone Joint Surg. 1992;74A:251–260.
27. Ravaglia G, Forti P, Maioli F, et al. Effect of micronutrient status on natural killer cell immune function in healthy free-living subjects aged ≥90 y. Am J Clin Nutr. 2000;71:590–598.
28. Saleh K, Olson M, Resig S, et al. Predictors of wound infection in hip and knee joint replacement: Results from a 20 year surveillance program. J Orthop Res. 2002;20:506–515.
29. Schmalzried TP, Amstutz HC, Au MK, Dorey FJ. Etiology of deep sepsis in total hip arthroplasty: The significance of hematogenous and recurrent infections. Clin Orthop. 1992;280:200–207.
30. Shenkin A. Micronutrients and outcome. Nutrition. 1997;13:825–828.
31. Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition among elderly hospitalized patients: A prospective study. JAMA. 1999;281:2013–2019.
32. Tenaud I, Sainte-Marie I, Jumbou O, Litoux P, Dreno B. In vitro modulation of keratinocyte wound healing integrins by zinc, copper and manganese. Br J Dermatol. 1999;140:26–34.
33. Thorsdottir I, Eriksen B, Eysteinsdottir S. Nutritional status at submission for dietetic services and screening for malnutrition at admission to hospital. Clin Nutr. 1999;18:15–21.
34. Vaxman F, Olender S, Maldonado H, et al. Variations of magnesium, iron, copper and zinc during the colonic wound healing process: Experimental study on rabbits. Eur Surg Res. 1992;24:283–290.
35. Wood RJ. Assessment of marginal zinc status in humans. J Nutr. 2000;130:1350–1354.
© 2004 Lippincott Williams & Wilkins, Inc.