Ir J Med Sci DOI 10.1007/s11845-014-1156-6
ORIGINAL ARTICLE
Surgery of the femur in HIV positive patients: a retrospective review from 2005 to 2011 F. Cummins • B. Ramasubbu • T. McCarthy C. Bergin • P. P. Grieve
•
Received: 16 November 2012 / Accepted: 3 June 2014 Ó Royal Academy of Medicine in Ireland 2014
Abstract Background There are an estimated 6,900 people with HIV living in Ireland. There is a significant prevalence of femoral osteonecrosis and risk factors for osteoporotic fractures. With this potential, increasing surgical workload, it is important to assess surgical demand and audit outcomes. Methods The hospital’s electronic records were examined. Between January 2005 and August 2011, 17 femoral surgeries were identified in seven patients. Patient records were retrospectively reviewed. Results Elective operations undertaken were eight hip replacements and one nail dynamisation. Eight emergency operations were undertaken: two DHS, one plating, one nailing, one revision nailing, two incision and drainages and one biopsy. All procedures were carried out using appropriate safety guidelines. Indications for surgery included femoral head osteonecrosis (n = 7), osteomyelitis (n = 3), proximal femoral fracture (n = 2), femoral shaft non-union (n = 1), dynamisation of a nail (n = 1), osteoarthritis (n = 1), fractured femur (n = 1), and revision nailing (n = 1). For two procedures the patient was not on highly active anti-retroviral treatment. All elective patients had CD4 counts greater than 200 pre-operatively. Six patients had undetectable viral loads. Of the eight emergency procedures, four procedures had no preoperative immune status recorded. Complications recorded were F. Cummins (&) B. Ramasubbu T. McCarthy P. P. Grieve Departments of Trauma and Orthopaedics, St James’s Hospital, Dublin, Ireland e-mail: [email protected] C. Bergin Department of GU Medicine and Infectious Disease (GUIDE), St James’s Hospital, Dublin, Ireland
three non-unions, one nail fracture, one lesser trochanter fracture and recurrence of osteomyelitis. No surgical site infections were recorded. Conclusions Complications were not related to immune status. The rate of surgical site infection in both elective and emergency procedures was low. The elective surgery patients can safely receive orthopaedic treatment in their regional orthopaedic unit. Due to the high non-infectious complication rates recorded in the emergency group, transfer to a tertiary facility with infectious disease expertise is advised. Keywords HIV Arthroplasty Surgical site infection Femoral fracture Irish Surgical complications
Introduction A recent publication by the health protection surveillance unit reported a total of 6,287 human immunodeficiency virus (HIV) infections to date, with 320 new cases diagnosed in Ireland in 2011 [1]. It is estimated that there are currently 6,900 people living with HIV in Ireland today [2]. HIV was associated with a disastrous prognosis, because of its association with opportunistic infections and malignancies. Though there is currently no effective, curative treatment for HIV infection, current anti-retroviral medications may provide a near normal lifespan [3]. The widespread use of highly active anti-retroviral therapy (HAART), has led to dramatically better outcomes for HIV infected patients [4]. Our unit follows the British HIV association’s guidelines for the initiation of HAART. Tables 1, 2 summarise these guidelines [5]. Unfortunately HAART is associated with serious complications, such as abnormalities in lipid metabolism,
123
Ir J Med Sci Table 1 BHIVA guidelines for initiating HAART
AIDS defining illness
Acute infection
Chronic infection
Yes
Yes
HIV related co-morbidity
Yes
Hepatitis B co-infection with CD4 \500cells/lL Hepatitis C co-infection with CD4 \500cells/lL
Yes
Non-AIDS defining malignancy requiring immunosuppressing chemotherapy or radiation
Yes
Hepatitis B co-infection with CD4 [500cells/lL if treating the hepatitis infection
Yes
Yes
AIDS defining infection or serious bacterial infection with CD4 \200cells/lL
Yes
Primary HIV infection with neurological involvement, CD4 \350cells/lL, AIDS defining illness
Yes
Reduce risk of transmission to partner if CD4 [350cells/lL
Yes
Yes
Table 2 Summary recommendations of choice of anti-retroviral therapy
NRTI backbonea Third agent
a
Preferred
Alternative
Tenofovir and emtricitabine
Abacavir and lamivudinea
Atazanavir/ritonavir
Lopinavir/ritonavir
Darunavir/ritonavir
Fosamprenavir/ritonavir
Efavirenz
Nevirapine
Raltegravir
Rilpivirine
Nucleos(t)ide reverse transcriptase inhibitor
endocrine disorders and effects on other major organ systems [6]. The use of HAART is also associated with loss of bone mineral density independent of HIV infection [7, 8]. Giardano et al. investigated BMD loss in patients receiving protease inhibitors. They suggest the decreased bone density may be due to a direct effect of protease inhibitors on bony remodelling or indirect effect on vitamin D metabolism [9]. Furthermore the initiation of HAART can trigger ‘Immune Reconstitution Inflammatory Syndrome’ (IRIS). This inflammatory cascade can decrease BMD and hence, place patients at an even greater risk of osteoporotic fracture [10]. The authors hypothesise that HAART use may have a bearing on adverse events and fracture risk in our patient population. With modern anti-retroviral therapy, an ageing HIV population is emerging [11]. This brings with it the normal
123
bone degenerative changes of age compounded by consequences of long-term HIV infection and the side-effects of long term anti-retroviral therapy. Decreasing BMD and osteoporosis secondary to disease and treatment has been well described [12], it is estimated that in the United States, by 2015, more than half of all HIV infections will be in people older than 50 years [13]. In Ireland, up to 10 % of new HIV infections are in the over 50 age group [14]. Sharma et al. [15] carried out a prospective study in a cohort of 230 HIV patients aged over 48 years of age, examining changes in bone mineral density (BMD). They discovered that patients with a diagnosis of AIDS, coupled with heroin abuse were at risk of losing bone density. Intravenous drug users (IDUs) are furthermore at an increased risk of fracture due to poor bone health induced by dietary deficits of calcium and vitamin D [16]. By auditing our past orthopaedic surgical experience with HIV infected patients, we hope to offer a guide to quantifying future surgical workload. Osteonecrosis of the femoral head appears to be an increasing common problem for people living with HIV, with rates of nearly 1 % reported [17]. Some authors have raised concerns regarding increased rates of surgical site infection (SSI) in HIV infected individuals. Drapeau et al. [18] carried out a multi-centre observational study of HIV patients undergoing surgery. 92.5 % of this cohort had their HIV infection under good control, but the cohort overall had a surgical site infection rate twice the noninfected populations’. This infection risk was not related to the immunological status of the patient. Abalo et al. [19] retrospectively analysed a cohort of HIV patients (n = 36) who underwent surgery for orthopaedic trauma. They noted high rates of surgical site infection (39 %) but risk of SSI was strongly related to HIV infection severity, decreasing CD4 counts and whether the initial fracture was open. Increasing long-term patient survival, prolonged antiviral therapy exposure and increasing co-morbidities will lead to an increasing need for elective hip arthroplasty. Also, decreasing bone mineral density, compounded with the increased risks of osteoporosis linked to HIV infection, HAART and heroin abuse, will also result in increasing numbers of HIV patients requiring emergency surgery for proximal femur fracture. With the concerns over SSI raised by other authors, and the potential future increased surgical workload, we carried out a retrospective review of our experience with proximal femur surgery in the HIV infected population in our unit.
Aims To investigate the outcomes of HIV positive patients’ undergoing femoral surgery in a large University Hospital in Ireland. The authors aimed to:
Ir J Med Sci Table 3 Patients’ CD4 counts and viral loads Surgery
Pre-Op CD4
Post-Op CD4
Pre-Op VL
Post-Op VL
Complications
1
Elec
613
724
2
Elec
724
656
Undetectable
90
Nil
90
Undetectable
3
Elec
868
Fracture of lesser trochanter
812
Undetectable
Undetectable
Nil
4
Elec
5
Elec
835
868
Undetectable
Undetectable
Nil
869
868
56
Undetectable
Nil
6 7
Elec
235
378
105, 185
779
Nil
Elec
402
400
Undetectable
Undetectable
8
Elec
211
9
Elec
400
402
Undetectable
Undetectable
Nil
10
Emer
431
667
Undetectable
Undetectable
Nil
11 12
Emer Emer
308
431 226
Undetectable
Undetectable Undetectable
Nil Non-union following DHS for intertrochanteric femoral fracture
13
Emer
Undetectable
Nil
14
Emer
15
Emer
16
Emer
226 400
Nil Nil
400
4,974
Undetectable
Non-union following ORIF for femoral shaft fracture
365
139,556
Periprosthetic and nail fracture
365
139,556
Non-union femur fracture following IM nailing
Highlight the potentially future increased likelihood of HIV patients, requiring surgery to the femur; To audit whether the patients were on appropriate HAART, whether elective or emergency, and to investigate if patients not on HAART had worse outcome(s); Investigate whether, surgical outcomes in HIV patients, are related to peri-operative immune status; Audit the incidence of surgical site infection (SSI).
Methods St James’s Hospital, Dublin houses the regional unit for treatment of infectious diseases and the national centre for hereditary coagulations disorders. Over 60 % of all HIV patients in Ireland receiving medical treatment attend St James’s Hospital. A significant number of HIV infected patients present with either femoral injury or symptomatic osteoarthritis secondary to femoral head osteonecrosis. Our institution covers a catchment area with high levels of social deprivation and intravenous drug use. The authors carried out a retrospective analysis of all HIV patients undergoing femoral surgery, to investigate complication rates and surgical outcomes. All patients who underwent proximal femur surgery with a diagnosis of HIV infection, between January 2005 and January 2011, were identified from the hospital inpatient discharge data (HIPE), an electronic record maintained by all acute Irish hospitals. The HIPE system also introduced codings for morbidity in January 2005. The HIPE system audits its data accuracy, with
140 validation checks routinely performed [20]. All HIV infections were confirmed with enzyme-linked immunosorbent assay (ELISA) and Western blotting. Patient medical records, laboratory test results and imaging findings were analysed retrospectively. Patient biographical data was recorded. Pre-operative and post-operative viral loads (HIV-RNA levels) and CD4 counts were used to assess patients’ immune status, the CD4 count considered a surrogate marker of immune status. A CD4 count of less than 200/lL was considered to be consistent with an epidemiological surveillance diagnosis of AIDS and high risk for opportunistic infection. HIV viral load, which is used to predict future changes in CD4? T cell count and disease progression risk, is also used as a marker of infectivity and immune status (Table 3). All procedures were carried out according to the infection control guidelines of our institution. Essential surgical implements only were used, with all sharps transferred in transit dish. Patient records (electronic or written) were marked as a transmission risk for viral disease and postoperatively all surgical instruments were tagged as a danger for transmission of blood borne viral disease for the safety of the sterilising personnel. The average post-operative follow-up was 25 months (range 1–68 months). Follow-up was standard for the procedure performed, for example THR patients were seen at 6 weeks and 3 months post procedure. The Department of GU Medicine and Infectious Diseases (GUIDE) offers a drop in outpatient service during office hours Monday to Friday, which allowed for a longer term follow-up for the patients. Blood borne markers of HIV disease activity and
123
Ir J Med Sci Table 4 Demographic data Gender
Number of surgeries
Risk factor for HIV acquisition
Age at operation
1
F
2
IVDU
2
F
5
Hetrosexual transmission
3
F
2
IVDU
27, 28
4
F
3
IVDU
29, 34, 34
5
M
1
Blood product
6
M
3
IVDU
7
M
2
IVDU
31, 31 39, 40, 40, 41, 42
On HAART at time of initial surgery?
HAART regimen
HCV
Yes
Ritonavir/darminavir/truvada
Lymphoma, HPV
Yes
Kivexa/Kaletra
HCV, ITP
No
HCV
Yes
Ritonavir/darminavir/truvada
HCV, FVIII D
Yes
Efavirenz/entricitabine/ tenotavir
26, 26
HCV
No
43, 43
HCV
No
45,
radiological tests were carried out, if clinically indicated, at these drop-in clinics.
Results Seven patients were identified, with a total of 17 operations on the femur. Nine elective and eight emergency procedures were identified. Four of the patients were female, three male. Average age was 35 years (range 26–43 years). For 13 of the 17 procedures, the patient was on HAART. Of the nine elective procedures, eight patients were on HAART and one commenced HIV therapy in the perioperative period. Of the eight emergency procedures, half the patients were not on HAART prior to their procedure. Of the seven patients, five acquired HIV infection from intravenous drug use, one from a contaminated blood product used to treat haemophilia and one from heterosexual contact (Table 4). Eight elective total hip replacements were performed for femoral head osteonecrosis, all of these patients were on HAART. One patient required three surgeries for primary femoral osteomyelitis, one patient required bilateral dynamic hip screws (1 year apart) for bilateral extra-capsular hip fractures. One patient required intramedullary nailing for a fractured femoral diaphysis due to a gunshot wound. He subsequently broke his intramedullary nail requiring revision intramedullary nailing within 1 month of his index surgery. The final trauma patient required open reduction and internal fixation for femoral fracture. Overall there was a non-statistically significant decreased correlation between complications and HAART use. (Correlation coefficient = -0.35, coefficient of determination = 0.12). However in the emergency surgery group HAART use was associated with decreased risk of surgical complication that approached but did not reach statistical significance.
123
Co-morbidities
(Correlation Coefficient = -0.55, coefficient of determination = 0.3). Looking at peri-operative immune status in the elective surgery group, pre-operative CD4 counts ranged from 211 to 869/ll. Average pre-procedure CD4 count was 573/ll. Preoperative CD4 counts were available for all elective patients. Post-operative CD4 counts ranged from 378 to 858/ll. Average post-operative CD4 count was 639/ll. Post-operative CD4 counts were not available for one patient who was 1 month post-procedure at follow-up. In the elective surgery group, pre-operative viral loads ranged from undetectable to 105, 185 copies/ml. The mode viral load for this group was undetectable. Post-operative viral loads ranged from undetectable to 779 copies/ml. There was no acute or late infectious post-operative complication in the elective surgery group. One complication was recorded in the elective surgery group. Looking at peri-operative immune status in the emergency surgery group, pre-operative CD4 counts ranged from 308 to 431/ll. Average pre-procedure CD4 count was 410/ll. Pre-operative CD4 counts were available for only half the emergency patients. Post-operative CD4 counts ranged from 226 to 667/ll. Average post-operative CD4 count was 418/ll. Post-operative CD4 counts were available for all procedures. In the emergency surgery group, pre-operative viral loads were available for only three of the procedures ranging from undetectable to 4,974 copies/ml. The mode viral load for this group was undetectable. Post-operative viral was undetectable for all but two procedures reflecting HAART started peri-operatively. There were multiple complications recorded in the emergency surgery group. One patient died of an AIDS related illness 18 months after his emergency surgery. The absolute risk of suffering a complication during elective surgery was 0.11. The absolute risk of suffering a
Ir J Med Sci Table 5 Correlations with complications Correlation co-efficient
Co-efficient of determination
Pre-operative CD4
-0.17
0.29
Post-operative CD4
-0.33
0.11
Pre-operative viral load
-0.15
0.02
Post-operative viral load
0.5
0.25
Emergency procedure
0.54
0.29
-0.35
0.12
0.94
0.88
HAART IDU
complication during emergency surgery was 0.625. Therefore, relative risk of a complication in emergency femoral surgery is approximately six times greater than for elective femoral surgery. Statistical calculations show no significant differences in the CD4 counts and viral loads when comparing the group who suffered surgical complications to the no complication group, thus suggesting, in our cohort, suffering a complication was not correlated with peri-operative immune status. There was strong, statistically significant correlation of suffering a complication and intravenous drug use. (Correlation coefficient = 0.94, coefficient of determination = 0.88). There was no correlation between suffering a complication and patient age (correlation coefficient = 0.29), duration of confirmed infection to surgery (correlation coefficient = 0.04), and hepatitis c virus co-infection (correlation coefficient = 0.24). There was a strong, statistically significant correlation between intravenous drug use and hepatitis C co-infection (Correlation coefficient = 0.87, coefficient of determination = 0.76). Complications recorded were three non-unions (one DHS, one diaphyseal nailing and one plating) and one nail fracture (Table 5). No surgical site infections were recorded. One elective THR suffered an intra-operative lesser trochanter fracture that was addressed at the time of surgery without any long-term sequelae.
Discussion This study, to audit the outcomes of HIV positive patients’ undergoing femoral surgery in a large University Hospital in Ireland, was undertaken to highlight the potentially future increased likelihood of HIV patients requiring surgery to the femur, audit prevalence of HAART in this patient cohort, to investigate whether surgical outcomes in HIV patients are related to peri-operative immune status and to audit the incidence of surgical site infection in this group. This study demonstrates the safety of elective hip arthroplasty in this high risk patient group with only one
complication recorded. No infectious complications were recorded in the whole group suggesting that with adequate control of HIV infection and appropriate perioperative management, the risk of surgical site infection is minimal. The majority of the patients led chaotic lifestyles, demonstrated by the high proportion of active or past intravenous drug use, but the results, especially for elective hip arthroplasty are encouraging and support offering THR for femoral head osteonecrosis. Due to the chaotic lifestyles of the emergency surgery cohort, demonstrated by the preoperative immune status being available in only half, and only half on HAART on presentation, a multidisciplinary approach in conjunction with an infectious disease service in a tertiary referral centre may be more appropriate after initial management of their orthopaedic injuries. The risk of suffering a complication significantly correlated with intravenous drug use, though this was not linked to immune status. The authors hypothesize that the increased complication rate in the IDUs is due to noncompliance with post-operative precautions as evidenced by the non-unions and the femoral nail fracture. Wieser et al. [21] report high infective complication rates post elective hip arthroplasty in a similar patient cohort, though noted infective complication was not related to HIV status. The one complication recorded during elective surgery was fracture of the lesser trochanter. This was not linked to HIV status and would be considered an uncommon intraoperative complication during hip arthroplasty. Limitations of this study include a small sample size, heterogeneity of the surgical procedures in the emergency surgery group, retrospective data collection and short follow-up for two of the surgeries. To the best of the authors’ knowledge, this is the largest cohort of patients living with HIV who underwent orthopaedic surgery, published in the Irish literature. These results reflect a small cohort of patients in a single centre. A further limitation to this study is that patients may have been surgically treated but not coded as having HIV infection, or may not have been yet diagnosed as being HIV positive. Our institution covers an area of high social deprivation and acts as a tertiary referral centre for the treatment of HIV infection, so is unlikely to match the demographics of a regional orthopaedic centre. Further, prospective analysis on a National scale with much larger cohort of patients, is recommended. In summary we believe this study demonstrates the complications recorded in both the elective and emergency surgery groups were not related to peri-operative immune status and infectious complications were low. Surgery for traumatic femoral injuries in HIV infected patients, though demonstrating increased complication rates generally, showed no increased risk of surgical site infection. The authors suggest that the elective surgery cohort can safely
123
Ir J Med Sci
have their joint replacements in regional orthopaedic units. Unfortunately the patients who underwent surgery for femoral fracture suffered non-infectious complications, possibly related to their lifestyles and/or perhaps poor health and nutritional status at the time of injury(s). Orthopaedic surgeons should expect to see increasing numbers of HIV patients requiring femoral surgery and it would be wise to develop further knowledge in this area and closer professional links with regional infectious disease services.
10.
11.
12.
13.
14.
References 15. 1. Health protection surveillance centre (2011). http://www.hpsc.ie/hpsc/ A-Z/HIVSTIs/HIVandAIDS/SurveillanceReports/File,13068,en.pdf 2. AVERTing HIV and AIDs. www.avert.org 3. Kiboneka A, Nyatia RJ, Nabiryo C et al (2009) Combination antiretroviral therapy in population affected by conflict: outcomes from large cohort in northern Uganda. BMJ 17(338):b201 4. Palella FJ Jr, Delaney KM, Moorman AC et al (1998) Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV outpatient study investigator. NEJM 338:853–860 5. Williams I, Churchill D, Anderson J et al (2012) British HIV Association guidelines for the treatment of HIV-1-positive adults with antiretroviral therapy 2012. HIV Med 13(Suppl 2):1–85 6. Carr A, Samaras K, Burton S et al (1998) A syndrome of peripheral lipodystrophy, hyperlipidemia, and insulin resistance in patients receiving HIV protease inhibitors. AIDS 12:F51–F52 7. Tebas P, Powderly WG, Claxton S et al (2000) Accelerated bone mineral loss in HIV-infected patients treated with potent antiretroviral therapy. AIDS 14:F63–F67 8. Moore AL, Vashisht A, Sabin CA et al (2001) Reduced bone density in HIV positive individuals. AIDS 15:1731–1733 9. Madeddu G, Spanu A, Solinas P et al (2004) Bone mass loss and vitamin D metabolism impairment in HIV patients receiving
123
16.
17.
18.
19.
20. 21.
highly active antiretroviral therapy. Q J Nucl Med Mol Imaging 48:39–48 Appay V, Sauce D (2008) Immune activation and inflammation in HIV-1 infection: causes and consequences. J Pathol 214: 231–241 Kearney F, Moore AR, Donegan CF et al (2010) The ageing of HIV: implications for geriatric medicine. Age Ageing 39(5):536–541 Epub 2010 Aug 4 Tebas P, Powderly WG, Claxton S et al (2000) Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy. AIDS 14(4):F63–F67 Medical management of older patients with HIV/AIDS (R01) (2010). [on-line]. Available at http://grants.nih.gov/grants/guide/ pa-files/PA-09-017.html. Accessed 14 Jan 2010 Smith RD, Delpech VC, Brown AE et al (2010) HIV transmission and high rates of late diagnoses among adults aged 50 years and over. AIDS 24(13):2109–2115 Sharma A, Flom PL, Weedon J et al (2010) Prospective study of bone mineral density changes in aging men with or at risk for HIV infection. AIDS 24(15):2337–2345 Estrella MM, Kirk GD, Mehta SH et al (2012) Vitamin D deficiency and persistent proteinuria among HIV-infected and uninfected injection drug users. AIDS 26(3):295–302 Yombi JC, Vandercam B, Wilmes D et al (2009) Osteonecrosis of the femoral head in patients with type 1 human immunodeficiency virus infection: clinical analysis and review. Clin Rheumatol 28(7):815–823 Drapeau CM, Pan A, Bellacossa C et al (2009) Surgical site infections in HIV-infected patients: results from an Italian prospective multicenter observational study. Infection 37(5): 455–460 Abalo A, Patassi A, James YE et al (2010) Risk factors for surgical wound infection in HIV-positive patients undergoing surgery for orthopaedic trauma. J Orthop Surg (Hong Kong) 18(2): 224–227 Wiley MM (2005) Using HIPE data as a research and planning tool: limitations and opportunities: a response. IJMS 174:2 Wieser K, Zinqq P, Betz M et al (2012) Total hip replacement in patients with history of illicit injecting drug use. Arch Orthop Trauma Surg 132(7):1037–1044
ORIGINAL ARTICLE
Surgery of the femur in HIV positive patients: a retrospective review from 2005 to 2011 F. Cummins • B. Ramasubbu • T. McCarthy C. Bergin • P. P. Grieve
•
Received: 16 November 2012 / Accepted: 3 June 2014 Ó Royal Academy of Medicine in Ireland 2014
Abstract Background There are an estimated 6,900 people with HIV living in Ireland. There is a significant prevalence of femoral osteonecrosis and risk factors for osteoporotic fractures. With this potential, increasing surgical workload, it is important to assess surgical demand and audit outcomes. Methods The hospital’s electronic records were examined. Between January 2005 and August 2011, 17 femoral surgeries were identified in seven patients. Patient records were retrospectively reviewed. Results Elective operations undertaken were eight hip replacements and one nail dynamisation. Eight emergency operations were undertaken: two DHS, one plating, one nailing, one revision nailing, two incision and drainages and one biopsy. All procedures were carried out using appropriate safety guidelines. Indications for surgery included femoral head osteonecrosis (n = 7), osteomyelitis (n = 3), proximal femoral fracture (n = 2), femoral shaft non-union (n = 1), dynamisation of a nail (n = 1), osteoarthritis (n = 1), fractured femur (n = 1), and revision nailing (n = 1). For two procedures the patient was not on highly active anti-retroviral treatment. All elective patients had CD4 counts greater than 200 pre-operatively. Six patients had undetectable viral loads. Of the eight emergency procedures, four procedures had no preoperative immune status recorded. Complications recorded were F. Cummins (&) B. Ramasubbu T. McCarthy P. P. Grieve Departments of Trauma and Orthopaedics, St James’s Hospital, Dublin, Ireland e-mail: [email protected] C. Bergin Department of GU Medicine and Infectious Disease (GUIDE), St James’s Hospital, Dublin, Ireland
three non-unions, one nail fracture, one lesser trochanter fracture and recurrence of osteomyelitis. No surgical site infections were recorded. Conclusions Complications were not related to immune status. The rate of surgical site infection in both elective and emergency procedures was low. The elective surgery patients can safely receive orthopaedic treatment in their regional orthopaedic unit. Due to the high non-infectious complication rates recorded in the emergency group, transfer to a tertiary facility with infectious disease expertise is advised. Keywords HIV Arthroplasty Surgical site infection Femoral fracture Irish Surgical complications
Introduction A recent publication by the health protection surveillance unit reported a total of 6,287 human immunodeficiency virus (HIV) infections to date, with 320 new cases diagnosed in Ireland in 2011 [1]. It is estimated that there are currently 6,900 people living with HIV in Ireland today [2]. HIV was associated with a disastrous prognosis, because of its association with opportunistic infections and malignancies. Though there is currently no effective, curative treatment for HIV infection, current anti-retroviral medications may provide a near normal lifespan [3]. The widespread use of highly active anti-retroviral therapy (HAART), has led to dramatically better outcomes for HIV infected patients [4]. Our unit follows the British HIV association’s guidelines for the initiation of HAART. Tables 1, 2 summarise these guidelines [5]. Unfortunately HAART is associated with serious complications, such as abnormalities in lipid metabolism,
123
Ir J Med Sci Table 1 BHIVA guidelines for initiating HAART
AIDS defining illness
Acute infection
Chronic infection
Yes
Yes
HIV related co-morbidity
Yes
Hepatitis B co-infection with CD4 \500cells/lL Hepatitis C co-infection with CD4 \500cells/lL
Yes
Non-AIDS defining malignancy requiring immunosuppressing chemotherapy or radiation
Yes
Hepatitis B co-infection with CD4 [500cells/lL if treating the hepatitis infection
Yes
Yes
AIDS defining infection or serious bacterial infection with CD4 \200cells/lL
Yes
Primary HIV infection with neurological involvement, CD4 \350cells/lL, AIDS defining illness
Yes
Reduce risk of transmission to partner if CD4 [350cells/lL
Yes
Yes
Table 2 Summary recommendations of choice of anti-retroviral therapy
NRTI backbonea Third agent
a
Preferred
Alternative
Tenofovir and emtricitabine
Abacavir and lamivudinea
Atazanavir/ritonavir
Lopinavir/ritonavir
Darunavir/ritonavir
Fosamprenavir/ritonavir
Efavirenz
Nevirapine
Raltegravir
Rilpivirine
Nucleos(t)ide reverse transcriptase inhibitor
endocrine disorders and effects on other major organ systems [6]. The use of HAART is also associated with loss of bone mineral density independent of HIV infection [7, 8]. Giardano et al. investigated BMD loss in patients receiving protease inhibitors. They suggest the decreased bone density may be due to a direct effect of protease inhibitors on bony remodelling or indirect effect on vitamin D metabolism [9]. Furthermore the initiation of HAART can trigger ‘Immune Reconstitution Inflammatory Syndrome’ (IRIS). This inflammatory cascade can decrease BMD and hence, place patients at an even greater risk of osteoporotic fracture [10]. The authors hypothesise that HAART use may have a bearing on adverse events and fracture risk in our patient population. With modern anti-retroviral therapy, an ageing HIV population is emerging [11]. This brings with it the normal
123
bone degenerative changes of age compounded by consequences of long-term HIV infection and the side-effects of long term anti-retroviral therapy. Decreasing BMD and osteoporosis secondary to disease and treatment has been well described [12], it is estimated that in the United States, by 2015, more than half of all HIV infections will be in people older than 50 years [13]. In Ireland, up to 10 % of new HIV infections are in the over 50 age group [14]. Sharma et al. [15] carried out a prospective study in a cohort of 230 HIV patients aged over 48 years of age, examining changes in bone mineral density (BMD). They discovered that patients with a diagnosis of AIDS, coupled with heroin abuse were at risk of losing bone density. Intravenous drug users (IDUs) are furthermore at an increased risk of fracture due to poor bone health induced by dietary deficits of calcium and vitamin D [16]. By auditing our past orthopaedic surgical experience with HIV infected patients, we hope to offer a guide to quantifying future surgical workload. Osteonecrosis of the femoral head appears to be an increasing common problem for people living with HIV, with rates of nearly 1 % reported [17]. Some authors have raised concerns regarding increased rates of surgical site infection (SSI) in HIV infected individuals. Drapeau et al. [18] carried out a multi-centre observational study of HIV patients undergoing surgery. 92.5 % of this cohort had their HIV infection under good control, but the cohort overall had a surgical site infection rate twice the noninfected populations’. This infection risk was not related to the immunological status of the patient. Abalo et al. [19] retrospectively analysed a cohort of HIV patients (n = 36) who underwent surgery for orthopaedic trauma. They noted high rates of surgical site infection (39 %) but risk of SSI was strongly related to HIV infection severity, decreasing CD4 counts and whether the initial fracture was open. Increasing long-term patient survival, prolonged antiviral therapy exposure and increasing co-morbidities will lead to an increasing need for elective hip arthroplasty. Also, decreasing bone mineral density, compounded with the increased risks of osteoporosis linked to HIV infection, HAART and heroin abuse, will also result in increasing numbers of HIV patients requiring emergency surgery for proximal femur fracture. With the concerns over SSI raised by other authors, and the potential future increased surgical workload, we carried out a retrospective review of our experience with proximal femur surgery in the HIV infected population in our unit.
Aims To investigate the outcomes of HIV positive patients’ undergoing femoral surgery in a large University Hospital in Ireland. The authors aimed to:
Ir J Med Sci Table 3 Patients’ CD4 counts and viral loads Surgery
Pre-Op CD4
Post-Op CD4
Pre-Op VL
Post-Op VL
Complications
1
Elec
613
724
2
Elec
724
656
Undetectable
90
Nil
90
Undetectable
3
Elec
868
Fracture of lesser trochanter
812
Undetectable
Undetectable
Nil
4
Elec
5
Elec
835
868
Undetectable
Undetectable
Nil
869
868
56
Undetectable
Nil
6 7
Elec
235
378
105, 185
779
Nil
Elec
402
400
Undetectable
Undetectable
8
Elec
211
9
Elec
400
402
Undetectable
Undetectable
Nil
10
Emer
431
667
Undetectable
Undetectable
Nil
11 12
Emer Emer
308
431 226
Undetectable
Undetectable Undetectable
Nil Non-union following DHS for intertrochanteric femoral fracture
13
Emer
Undetectable
Nil
14
Emer
15
Emer
16
Emer
226 400
Nil Nil
400
4,974
Undetectable
Non-union following ORIF for femoral shaft fracture
365
139,556
Periprosthetic and nail fracture
365
139,556
Non-union femur fracture following IM nailing
Highlight the potentially future increased likelihood of HIV patients, requiring surgery to the femur; To audit whether the patients were on appropriate HAART, whether elective or emergency, and to investigate if patients not on HAART had worse outcome(s); Investigate whether, surgical outcomes in HIV patients, are related to peri-operative immune status; Audit the incidence of surgical site infection (SSI).
Methods St James’s Hospital, Dublin houses the regional unit for treatment of infectious diseases and the national centre for hereditary coagulations disorders. Over 60 % of all HIV patients in Ireland receiving medical treatment attend St James’s Hospital. A significant number of HIV infected patients present with either femoral injury or symptomatic osteoarthritis secondary to femoral head osteonecrosis. Our institution covers a catchment area with high levels of social deprivation and intravenous drug use. The authors carried out a retrospective analysis of all HIV patients undergoing femoral surgery, to investigate complication rates and surgical outcomes. All patients who underwent proximal femur surgery with a diagnosis of HIV infection, between January 2005 and January 2011, were identified from the hospital inpatient discharge data (HIPE), an electronic record maintained by all acute Irish hospitals. The HIPE system also introduced codings for morbidity in January 2005. The HIPE system audits its data accuracy, with
140 validation checks routinely performed [20]. All HIV infections were confirmed with enzyme-linked immunosorbent assay (ELISA) and Western blotting. Patient medical records, laboratory test results and imaging findings were analysed retrospectively. Patient biographical data was recorded. Pre-operative and post-operative viral loads (HIV-RNA levels) and CD4 counts were used to assess patients’ immune status, the CD4 count considered a surrogate marker of immune status. A CD4 count of less than 200/lL was considered to be consistent with an epidemiological surveillance diagnosis of AIDS and high risk for opportunistic infection. HIV viral load, which is used to predict future changes in CD4? T cell count and disease progression risk, is also used as a marker of infectivity and immune status (Table 3). All procedures were carried out according to the infection control guidelines of our institution. Essential surgical implements only were used, with all sharps transferred in transit dish. Patient records (electronic or written) were marked as a transmission risk for viral disease and postoperatively all surgical instruments were tagged as a danger for transmission of blood borne viral disease for the safety of the sterilising personnel. The average post-operative follow-up was 25 months (range 1–68 months). Follow-up was standard for the procedure performed, for example THR patients were seen at 6 weeks and 3 months post procedure. The Department of GU Medicine and Infectious Diseases (GUIDE) offers a drop in outpatient service during office hours Monday to Friday, which allowed for a longer term follow-up for the patients. Blood borne markers of HIV disease activity and
123
Ir J Med Sci Table 4 Demographic data Gender
Number of surgeries
Risk factor for HIV acquisition
Age at operation
1
F
2
IVDU
2
F
5
Hetrosexual transmission
3
F
2
IVDU
27, 28
4
F
3
IVDU
29, 34, 34
5
M
1
Blood product
6
M
3
IVDU
7
M
2
IVDU
31, 31 39, 40, 40, 41, 42
On HAART at time of initial surgery?
HAART regimen
HCV
Yes
Ritonavir/darminavir/truvada
Lymphoma, HPV
Yes
Kivexa/Kaletra
HCV, ITP
No
HCV
Yes
Ritonavir/darminavir/truvada
HCV, FVIII D
Yes
Efavirenz/entricitabine/ tenotavir
26, 26
HCV
No
43, 43
HCV
No
45,
radiological tests were carried out, if clinically indicated, at these drop-in clinics.
Results Seven patients were identified, with a total of 17 operations on the femur. Nine elective and eight emergency procedures were identified. Four of the patients were female, three male. Average age was 35 years (range 26–43 years). For 13 of the 17 procedures, the patient was on HAART. Of the nine elective procedures, eight patients were on HAART and one commenced HIV therapy in the perioperative period. Of the eight emergency procedures, half the patients were not on HAART prior to their procedure. Of the seven patients, five acquired HIV infection from intravenous drug use, one from a contaminated blood product used to treat haemophilia and one from heterosexual contact (Table 4). Eight elective total hip replacements were performed for femoral head osteonecrosis, all of these patients were on HAART. One patient required three surgeries for primary femoral osteomyelitis, one patient required bilateral dynamic hip screws (1 year apart) for bilateral extra-capsular hip fractures. One patient required intramedullary nailing for a fractured femoral diaphysis due to a gunshot wound. He subsequently broke his intramedullary nail requiring revision intramedullary nailing within 1 month of his index surgery. The final trauma patient required open reduction and internal fixation for femoral fracture. Overall there was a non-statistically significant decreased correlation between complications and HAART use. (Correlation coefficient = -0.35, coefficient of determination = 0.12). However in the emergency surgery group HAART use was associated with decreased risk of surgical complication that approached but did not reach statistical significance.
123
Co-morbidities
(Correlation Coefficient = -0.55, coefficient of determination = 0.3). Looking at peri-operative immune status in the elective surgery group, pre-operative CD4 counts ranged from 211 to 869/ll. Average pre-procedure CD4 count was 573/ll. Preoperative CD4 counts were available for all elective patients. Post-operative CD4 counts ranged from 378 to 858/ll. Average post-operative CD4 count was 639/ll. Post-operative CD4 counts were not available for one patient who was 1 month post-procedure at follow-up. In the elective surgery group, pre-operative viral loads ranged from undetectable to 105, 185 copies/ml. The mode viral load for this group was undetectable. Post-operative viral loads ranged from undetectable to 779 copies/ml. There was no acute or late infectious post-operative complication in the elective surgery group. One complication was recorded in the elective surgery group. Looking at peri-operative immune status in the emergency surgery group, pre-operative CD4 counts ranged from 308 to 431/ll. Average pre-procedure CD4 count was 410/ll. Pre-operative CD4 counts were available for only half the emergency patients. Post-operative CD4 counts ranged from 226 to 667/ll. Average post-operative CD4 count was 418/ll. Post-operative CD4 counts were available for all procedures. In the emergency surgery group, pre-operative viral loads were available for only three of the procedures ranging from undetectable to 4,974 copies/ml. The mode viral load for this group was undetectable. Post-operative viral was undetectable for all but two procedures reflecting HAART started peri-operatively. There were multiple complications recorded in the emergency surgery group. One patient died of an AIDS related illness 18 months after his emergency surgery. The absolute risk of suffering a complication during elective surgery was 0.11. The absolute risk of suffering a
Ir J Med Sci Table 5 Correlations with complications Correlation co-efficient
Co-efficient of determination
Pre-operative CD4
-0.17
0.29
Post-operative CD4
-0.33
0.11
Pre-operative viral load
-0.15
0.02
Post-operative viral load
0.5
0.25
Emergency procedure
0.54
0.29
-0.35
0.12
0.94
0.88
HAART IDU
complication during emergency surgery was 0.625. Therefore, relative risk of a complication in emergency femoral surgery is approximately six times greater than for elective femoral surgery. Statistical calculations show no significant differences in the CD4 counts and viral loads when comparing the group who suffered surgical complications to the no complication group, thus suggesting, in our cohort, suffering a complication was not correlated with peri-operative immune status. There was strong, statistically significant correlation of suffering a complication and intravenous drug use. (Correlation coefficient = 0.94, coefficient of determination = 0.88). There was no correlation between suffering a complication and patient age (correlation coefficient = 0.29), duration of confirmed infection to surgery (correlation coefficient = 0.04), and hepatitis c virus co-infection (correlation coefficient = 0.24). There was a strong, statistically significant correlation between intravenous drug use and hepatitis C co-infection (Correlation coefficient = 0.87, coefficient of determination = 0.76). Complications recorded were three non-unions (one DHS, one diaphyseal nailing and one plating) and one nail fracture (Table 5). No surgical site infections were recorded. One elective THR suffered an intra-operative lesser trochanter fracture that was addressed at the time of surgery without any long-term sequelae.
Discussion This study, to audit the outcomes of HIV positive patients’ undergoing femoral surgery in a large University Hospital in Ireland, was undertaken to highlight the potentially future increased likelihood of HIV patients requiring surgery to the femur, audit prevalence of HAART in this patient cohort, to investigate whether surgical outcomes in HIV patients are related to peri-operative immune status and to audit the incidence of surgical site infection in this group. This study demonstrates the safety of elective hip arthroplasty in this high risk patient group with only one
complication recorded. No infectious complications were recorded in the whole group suggesting that with adequate control of HIV infection and appropriate perioperative management, the risk of surgical site infection is minimal. The majority of the patients led chaotic lifestyles, demonstrated by the high proportion of active or past intravenous drug use, but the results, especially for elective hip arthroplasty are encouraging and support offering THR for femoral head osteonecrosis. Due to the chaotic lifestyles of the emergency surgery cohort, demonstrated by the preoperative immune status being available in only half, and only half on HAART on presentation, a multidisciplinary approach in conjunction with an infectious disease service in a tertiary referral centre may be more appropriate after initial management of their orthopaedic injuries. The risk of suffering a complication significantly correlated with intravenous drug use, though this was not linked to immune status. The authors hypothesize that the increased complication rate in the IDUs is due to noncompliance with post-operative precautions as evidenced by the non-unions and the femoral nail fracture. Wieser et al. [21] report high infective complication rates post elective hip arthroplasty in a similar patient cohort, though noted infective complication was not related to HIV status. The one complication recorded during elective surgery was fracture of the lesser trochanter. This was not linked to HIV status and would be considered an uncommon intraoperative complication during hip arthroplasty. Limitations of this study include a small sample size, heterogeneity of the surgical procedures in the emergency surgery group, retrospective data collection and short follow-up for two of the surgeries. To the best of the authors’ knowledge, this is the largest cohort of patients living with HIV who underwent orthopaedic surgery, published in the Irish literature. These results reflect a small cohort of patients in a single centre. A further limitation to this study is that patients may have been surgically treated but not coded as having HIV infection, or may not have been yet diagnosed as being HIV positive. Our institution covers an area of high social deprivation and acts as a tertiary referral centre for the treatment of HIV infection, so is unlikely to match the demographics of a regional orthopaedic centre. Further, prospective analysis on a National scale with much larger cohort of patients, is recommended. In summary we believe this study demonstrates the complications recorded in both the elective and emergency surgery groups were not related to peri-operative immune status and infectious complications were low. Surgery for traumatic femoral injuries in HIV infected patients, though demonstrating increased complication rates generally, showed no increased risk of surgical site infection. The authors suggest that the elective surgery cohort can safely
123
Ir J Med Sci
have their joint replacements in regional orthopaedic units. Unfortunately the patients who underwent surgery for femoral fracture suffered non-infectious complications, possibly related to their lifestyles and/or perhaps poor health and nutritional status at the time of injury(s). Orthopaedic surgeons should expect to see increasing numbers of HIV patients requiring femoral surgery and it would be wise to develop further knowledge in this area and closer professional links with regional infectious disease services.
10.
11.
12.
13.
14.
References 15. 1. Health protection surveillance centre (2011). http://www.hpsc.ie/hpsc/ A-Z/HIVSTIs/HIVandAIDS/SurveillanceReports/File,13068,en.pdf 2. AVERTing HIV and AIDs. www.avert.org 3. Kiboneka A, Nyatia RJ, Nabiryo C et al (2009) Combination antiretroviral therapy in population affected by conflict: outcomes from large cohort in northern Uganda. BMJ 17(338):b201 4. Palella FJ Jr, Delaney KM, Moorman AC et al (1998) Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV outpatient study investigator. NEJM 338:853–860 5. Williams I, Churchill D, Anderson J et al (2012) British HIV Association guidelines for the treatment of HIV-1-positive adults with antiretroviral therapy 2012. HIV Med 13(Suppl 2):1–85 6. Carr A, Samaras K, Burton S et al (1998) A syndrome of peripheral lipodystrophy, hyperlipidemia, and insulin resistance in patients receiving HIV protease inhibitors. AIDS 12:F51–F52 7. Tebas P, Powderly WG, Claxton S et al (2000) Accelerated bone mineral loss in HIV-infected patients treated with potent antiretroviral therapy. AIDS 14:F63–F67 8. Moore AL, Vashisht A, Sabin CA et al (2001) Reduced bone density in HIV positive individuals. AIDS 15:1731–1733 9. Madeddu G, Spanu A, Solinas P et al (2004) Bone mass loss and vitamin D metabolism impairment in HIV patients receiving
123
16.
17.
18.
19.
20. 21.
highly active antiretroviral therapy. Q J Nucl Med Mol Imaging 48:39–48 Appay V, Sauce D (2008) Immune activation and inflammation in HIV-1 infection: causes and consequences. J Pathol 214: 231–241 Kearney F, Moore AR, Donegan CF et al (2010) The ageing of HIV: implications for geriatric medicine. Age Ageing 39(5):536–541 Epub 2010 Aug 4 Tebas P, Powderly WG, Claxton S et al (2000) Accelerated bone mineral loss in HIV-infected patients receiving potent antiretroviral therapy. AIDS 14(4):F63–F67 Medical management of older patients with HIV/AIDS (R01) (2010). [on-line]. Available at http://grants.nih.gov/grants/guide/ pa-files/PA-09-017.html. Accessed 14 Jan 2010 Smith RD, Delpech VC, Brown AE et al (2010) HIV transmission and high rates of late diagnoses among adults aged 50 years and over. AIDS 24(13):2109–2115 Sharma A, Flom PL, Weedon J et al (2010) Prospective study of bone mineral density changes in aging men with or at risk for HIV infection. AIDS 24(15):2337–2345 Estrella MM, Kirk GD, Mehta SH et al (2012) Vitamin D deficiency and persistent proteinuria among HIV-infected and uninfected injection drug users. AIDS 26(3):295–302 Yombi JC, Vandercam B, Wilmes D et al (2009) Osteonecrosis of the femoral head in patients with type 1 human immunodeficiency virus infection: clinical analysis and review. Clin Rheumatol 28(7):815–823 Drapeau CM, Pan A, Bellacossa C et al (2009) Surgical site infections in HIV-infected patients: results from an Italian prospective multicenter observational study. Infection 37(5): 455–460 Abalo A, Patassi A, James YE et al (2010) Risk factors for surgical wound infection in HIV-positive patients undergoing surgery for orthopaedic trauma. J Orthop Surg (Hong Kong) 18(2): 224–227 Wiley MM (2005) Using HIPE data as a research and planning tool: limitations and opportunities: a response. IJMS 174:2 Wieser K, Zinqq P, Betz M et al (2012) Total hip replacement in patients with history of illicit injecting drug use. Arch Orthop Trauma Surg 132(7):1037–1044
