Japanese Journal of Clinical Oncology, 2020, 50(8)867–872 doi: 10.1093/jjco/hyaa060 Advance Access Publication Date: 15 May 2020 Original Article

Original Article Study on the prevention of infection in terminal cancer patients applying epidural analgesia by adding cefazolin to anesthetics: a randomized Downloaded from https://academic.oup.com/jjco/article/50/8/867/5837331 by guest on 01 October 2021 controlled trial Pengcheng Xie*, Zhanfang Li*, Jingli Yang and Yiming Wu

Department of Anesthesiology, Pudong Hospital, Pudong Medical Center, Pudong, Shanghai,

*For reprints and all correspondence: Pengcheng Xie, Department of Anesthesiology, Shanghai Pudong Hospital, Fudan University Pudong Medical Center, 2800 Gongwei Road, Huinan Town, Pudong, Shanghai 201399, China. E-mail: [email protected] and Zhanfang Li. E-mail: [email protected]

Received 1 March 2020; Editorial Decision 13 April 2020; ; Accepted 14 April 2020

Abstract Objective: To observe the antibacterial effect of adding cefazolin into anesthetics in patients with terminal cancer undergoing long-term epidural analgesia. Methods: Patients undergoing epidural analgesia with terminal cancer were randomly divided into two groups: the conventional drug group (group C) and the cefazolin group (group G). Both groups were given levobupivacaine and morphine, while cefazolin was added to group G. The mean arterial pressure (MAP), heart rate (HR), respiratory rates (R), visual analogue scale (VAS) scores, satisfaction and complications of patients in the two groups were observed. 3 ml of the used analgesic was taken for bacterial culture when replacing the new analgesic case. Results: HR, MAP, R, VAS scores, dosages of morphine, satisfaction of the analgesic and the complications were not significantly different between the two groups (P > 0.05). The overall satisfaction of patients in group G was significantly higher than that in group C (P < 0.05). One patient’s anesthetics in group C were infected with colibacillus. Four patients in group C got infected in their epidural puncture sites. There was no infection in group G (P < 0.05). Conclusion: Adding cefazolin to local anesthetics could effectively prevent bacterial infection and ensured the safety of epidural analgesia for patients with terminal cancer.

Key words: anesthetics, patients with cancer, epidural analgesia, infection

Introduction the life quality of patients, and effective analgesic treatment is Factors such as high work pressure, unhealthy diet, irregular working particularly important to improvement of the life quality of patients hours and polluted environment contribute to the increasing number with terminal cancers. Medical workers have been trying their best of cancer patients. There are many complications in patients with to alleviate the pain of patients for a long time, so as to improve the terminal cancers, among which severe pains are undoubtedly the life quality of patients with terminal cancers (2–4). biggest factor reducing the life quality of patients. Studies have shown With the popularization of patient-controlled analgesia (PCA), that 75% of cancer patients experienced varying degrees of pains, epidural analgesia with high efficiency and low complications had with 40–50% experiencing severe pains and 25–30% experiencing attracted more and more attention of patients with terminal cancer. intolerable pains (1). Cancer pain is an important factor affecting This required large doses of epidural anesthetics, but the sterility

© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: [email protected] 867 868 Epidural analgesia for terminal cancer patients and stability of the compound analgesic solution limit its extensive province, China) was added to group G and none to group C. These clinical application in cancer analgesia. In this study, researchers analgesics could support the treatment for 4–5 days. Analgesics were added cefazolin to epidural analgesics to observe the effectiveness refilled by a doctor applying the same recipe. Infusion pump usage: in preventing bacterial infection. first dose 2 ml, background dose 2 ml·h−1, single additional dose 0.5 ml, 15 min automatic lockup and maximum output 4 ml·h−1. The case was replaced when the liquid was about to run out. 3ml of Materials and methods the left liquid was sent for bacterial culture. Participants Patients with VAS scores higher than 4 were treated with 5 mg of morphine orally. All patients were treated with no other analgesics This study was a prospective, randomized, comparative and double- but antitumor drugs. If the patients had intolerable complications, blind controlled trial. It has been reported in line with the Consol- the amount of epidural anesthetics should be reduced or even dis- idated Standards of Reporting Trials (CONSORT 2010) guidelines. continued. And the study should be discontinued. The study protocol was approved by the Human Research Ethics Duration of the catheter placement: the catheter was retained Committee and registered on August 2018. This study followed for life unless otherwise specified. The catheter was immediately

standardized ethical and safety principles of Good Clinical Practice Downloaded from https://academic.oup.com/jjco/article/50/8/867/5837331 by guest on 01 October 2021 removed if the epidural space infection or the epidural catheter guidelines and the declaration of Helsinki. Based on written informed strayed into the subarachnoid space or the epidural catheter came consents, 127 patients with terminal cancers (patients with TNM out. staging III and IV) were enrolled in this study between August 2018 All patients were hospitalized for 3 days to observe the analgesic and April 2019. Patients were allocated randomly according to a effect. They could choose to be hospitalized or discharged to continue computer-generated sequence of random numbers, to undergo with treatment if there was no complications. The patients’ conditions conventional drugs (group C with 63 patients) or cefazolin (group and parameters were followed up by the residents. Symptomatic G with 64 patients). Drug resistance had developed in oral and treatment was performed, and the disease was carefully observed if intravenous opioid analgesics for these patients. Inclusion criteria: complications occurred. ASA I–III; patients with abdominal, pelvic or lower limbs’ cancer; aged 30–75 year old; and being able to fill up VAS score table. Exclusion criteria: hypertension patients with poor control, severe arrhythmia, poor function of the heart and lung, acid–base or elec- Data records trolyte disturbance, severe anemia, spinal metastasis, smoking his- The patients’ basic vital signs were recorded, and 3 ml of anesthetics tory, psychotropic drugs abuse, history of mental illness, infection at was extracted for bacterial culture. MAP, HR, R, VAS scores and the puncture site, allergy to cefazolin and steroid treatment. complications were recorded every 15 min for 2 h after puncture finished. Patients’ vital signs and VAS scores were recorded on day 2, Paracentesis 4, 6, 8, 10, 15, 20, 25, 30 and 35 after the treatment. The total All epidural catheterization procedures were performed by the same oral dose of morphine was recorded. Complications were observed: two trained anesthetists who were unaware of the patient’s ran- urinary retention, cutaneous pruritus, rashes, nausea and vomiting, domization. The residents who did not participate in the study dizzy, headache, dyspnea and constipation. The satisfaction of the made records of the data during and post-epidural puncture. On analgesic, overall satisfaction and total number of infections were the day of the paracentesis, patients were delivered to the operating recorded. 3 ml of the used analgesic was taken for bacterial culture room at 8.30 a.m. Heart rate (three-lead ECG), noninvasive arterial when replacing the new analgesic pump case. pressure and oxygen saturation were continuously monitored during Overall satisfaction: patients were distributed with questionnaire −1 the procedure; supplemental oxygen (2 l min ) was administered to comment on the doctors’ service attitude, quality of answers, via nasal prongs. explanation of particular diseases, professionalism, privacy protec- All patients were placed in a lateral position and received punc- tion and treatment effectiveness on complications. Each item was ture needle through the median of L2–L3 space. 5 ml of 1% lidocaine scored on a 10-point scale, aggregated to 60 points at maximum. was injected after the epidural catheterization was successful. The If the patient’s anesthetics got infected, epidural catheter should catheter was placed into epidural space for 4 cm, and a subcutaneous be removed, and the study should be terminated. The patients were 5 cm tunnel was established towards the cephalic side. The epidural given the appropriate antibiotic orally in a leukocytosis situation, catheter was pulled and connected to the bacterial filter and the under careful monitoring for fever and infection in epidural space. analgesic case in order (Apon medical company, Apon filter and Apon analgesic case, , province, China). Sensory regression was measured by blunt needles 10 min later.The block was considered successful when the anesthesia level was between T8 and Statistical analysis S5. When the subarachnoid anesthesia occurred or the block failed, The primary endpoint was none of the patients were infected on day the patients were repunctured. 35 after epidural catheterization. On the basis of previous studies assuming a significant difference of 8% in the incidence of patients with infection on day 35 after epidural analgesia with a type I Study design error of 0.05 and a power of 0.8, a sample size of 60 patients per The drug’s composition: 50 ml of 0.75% levobupivacaine (Hen- group was required. We decided to include 127 patients to take into grui Medicine, Suka, , Jiangsu province, China) and account the possible loss of follow-up at the 35-day evaluation. Data 5 mg of morphine (Northeast Pharm, morphine, , Liaoning were analyzed using the Statistical Package for the Social Science province, China) were mixed and diluted with normal saline to version (SPSS for Windows, release 20.0; SPSS, Chicago, IL, USA). 250 ml. 0.5 g cefazolin (CR Sanjiu, Xintailin, , Guangdong A Kolmogorov–Smirnov test was used, and stratified distribution Jpn J Clin Oncol, 2020, Vol. 50, No. 8 869

Ta b l e 1. Comparison of baseline characteristics between groups

Parameters Group C Group G P value

Age (years) 58.31 ± 15.27 56.58 ± 16.83 0.661 Weight (kg) 52.46 ± 11.28 55.39 ± 13.64 0.342 Gender (female/male) 33/30 35/29 0.860 ASA scores (I/II/III) 0/29/34 0/27/37 0.722 Diabetes mellitus 15 17 0.839 Pre-analgesia HR (beats/min) 79.33 ± 12.58 76.54 ± 15.42 0.421 Pre-analgesia MAP (mmHg) 82.09 ± 16.75 84.73 ± 14.38 0.491 Pre-analgesia R (beats/min) 18.32 ± 2.08 19.17 ± 1.83 0.080 Pre-analgesia VAS scores 6.80 ± 1.84 6.94 ± 1.55 0.737

Data shown as mean ± standard deviation or number of patients, as indicated. Downloaded from https://academic.oup.com/jjco/article/50/8/867/5837331 by guest on 01 October 2021 plots were examined to verify the normality of distribution of con- Ta b l e 2 . Comparison of cancer types between groups tinuous variables. Baseline characteristics (gender, age, ASA scores, preoperative VAS scores and types of cancers) were compared across Cancer types Group C Group G P value treatment groups using a two-way analysis of variance (ANOVA) or Gastric cancer 9 13 0.681 Fisher’s exact test. The total amount of oral morphine was analyzed Colorectal cancer 15 12 by using an ANOVA. Repeated measures (MAP, HR, R, VAS scores) Prostatic cancer 13 9 were evaluated using two-way ANOVA and after epidural puncture Bladder cancer 9 11 comparisons at various times using Bonferroni’s type I error correc- Ovarian cancer 3 6 tion for multiple comparisons. Quantitative variables were expressed Endometrial cancer 4 7 as mean (SD). Side-effects, bacterial culture and patient satisfaction Cervical cancer 8 5 were assessed using Fisher’s exact test. No interim analysis was Lower limbs’ cancer 2 1 planned during the study, and P < 0.05 was set as the threshold ± for statistical significance. All graphics were made by GraphPad Data shown as mean standard deviation or number of patients, as indicated. Prism 7.

Results The primary outcomes were the number of infected patients. The sec- ondary outcomes were the satisfaction of patients, the hemodynamic changes, the effects of analgesic, the dosage of remedial analgesics and the complications. There were no patients having subarachnoid anesthesia or intol- erable complications in both groups during the 35 days. Three patients died in group C on day 19, 22 and 31, respectively. Two patients died in group G on day 15 and 27. One patient in group G got epidural catheter felling off on day 17 who re-performed with epidural puncture. The epidural catheter was removed from one patient in group C who had leukocytosis and anesthetic infection on day 26. The patient took cefuroxime orally and did not appear to have fever or epidural infection. Four patients in group C got effected with their epidural puncture sites on day 11, 19, 23 and 27, respectively. They were cured after local skin disinfection. Data from 127 patients were analyzed: 63 patients in group C and 64 patients in group G. The two groups of patients were comparable in terms of baseline characteristics data, i.e. age, weight, gender, ASA scores, basic hemodynamic parameters (HR, MAP,R) and VAS scores Figure 1. Comparison of HR between the two groups. (P > 0.05) (Table 1). The 127 patients had abdominal, pelvic or lower limbs can- cers. There was no statistically significant difference in cancer types between the two groups (P > 0.05) (Table 2). groups (P > 0.05). The overall satisfaction of patients in group There was no significant difference in HR, MAP, R and G was significantly higher than that in group C (P < 0.05). The VAS scores between the two groups during 35 days (P > 0.05) incidence of infection was significantly different between the two (Figs 1–4). groups (P < 0.05): one patient in group C got analgesic infection, There was no significant difference in complications, dosages and four had infection at the puncture site, while none in group G of morphine or satisfaction with analgesic effects between the two Table 3. 870 Epidural analgesia for terminal cancer patients

Ta b l e 3 . Comparison of complications, dosages of morphine and satisfaction of the analgesic between groups

Parameters Group C Group G P value

Mild Moderate Severe Mild Moderate Severe

Urinary retention 1 1 2 1 0.634 Cutaneous pruritus 2 1 3 2 Rashes 0 0 Nausea and vomiting 2 1 Dizzy 1 0 Headache 0 0 Dyspnea 0 0 Constipation 1 2 1 Dosages of morphine 195 225 0.530

(mg) Downloaded from https://academic.oup.com/jjco/article/50/8/867/5837331 by guest on 01 October 2021 Satisfaction of the 51 55 0.483 analgesic Overall satisfaction 46.76 ± 5.89 50.46 ± 7.52 0.003 Total number of 5 0 0.028 infections

Figure 2. Comparison of MAP between the two groups. Figure 3. Comparison of R between the two groups.

high incidence of nausea and vomiting, dizziness and even respiratory Discussion depression (8,9). Epidural analgesia is developed on the basis of Patients with terminal cancer are all accompanied by severe pains, epidural anesthesia and having been a commonly treatment for which seriously affect their sleep, daily life and social life resulting refractory cancer pains in the world. Due to pain and anxiety, cancer in a serious decline in their quality of life (5). The treatment of patients have increasing sympathetic excitability, and their breathing pain in cancer patients has become an important challenge for heart rates and blood pressures increased significantly when suffering clinicians. The WHO also put analgesia for cancer patients as one from pains. Appropriate analgesic effect will help improving the of cancer treatment focuses. Many studies have proved that epidural mood of patients and their life quality (10). Epidural analgesic analgesia is effective in patients with terminal cancers (6,7), and this treatment can bring good analgesic effect fast and effectively, study further explored the safety of anesthetics applied in long-term significantly reducing the complications of analgesic drugs, which epidural analgesia. are more and more used in patients with terminal cancers (11,12). Oral analgesics are not suitable for long-term analgesia of In this study, continuous supply of drugs through epidural space patients with terminal cancer pains due to their slow onset and combined with patient-controlled analgesia can effectively maintain weak analgesic intensity. Although intravenous analgesics bring stable drug concentration with satisfactory analgesic effect. Long- strong analgesic intensity, they are not suitable for long-term use term application did not decrease analgesic efficacy, showing no drug and not working for patients to use at home due to inconvenience in resistance, addiction or serious complications. Therefore, it can meet application, peak blood concentration, drug tolerance and addiction, the need of long-term treatment on cancer pains. Jpn J Clin Oncol, 2020, Vol. 50, No. 8 871

little systemic influence and few complications while being able to effectively prevent analgesic infection. Although the result needs to be confirmed in a large sample size, it is an effective, convenient and safe analgesic method (15).

Conflict of interest statement The authors declare that they have no conflict of interest.

Declarations The study was approved by the Human Research Ethics Committee of the Shanghai Pudong Hospital (No. LX2018-001). The trial registration identifier at clinicaltrials.gov is ChiCTR1800017577. Downloaded from https://academic.oup.com/jjco/article/50/8/867/5837331 by guest on 01 October 2021

Authors’ responsibility All authors confirm: We have not previously published or have not submitted the same manuscript elsewhere. We took a significant part in the work and approved the final version of the manuscript. We have complied with ethical standards. We agree to grant Oxford Figure 4. Comparison of VAS scores between the two groups. University Press a license to publish the accepted article when the manuscript is accepted. We have obtained all necessary permissions to publish any figures or tables in the manuscript and assure that In this study, urinary retention occurred in five patients, all of the authors will pay for any necessary charges. We have informed whom were elderly men with prostatic hypertrophy. Their symptoms all of the persons named in the acknowledgments of papers of their were improved after 3–4 days of catheterization. Eight patients had inclusion in this section. cutaneous pruritus due to taking morphine orally. Three patients got nausea and vomiting and one being dizzy likely caused by hypovolemia due to decreasing diet. The symptoms were mitigated Authors’ contributions with diet improvement. There was no patient having rashes, headache Conceptualization, Zhanfang Li; methodology, Jingli Yang; formal or dyspnea, showing the advantages of epidural analgesia. analysis and investigation, Yiming Wu; writing—original draft One of the key aspects of epidural analgesia is the prevention preparation, Pengcheng Xie; writing—review and editing, Pengcheng of infection (13). Chen’s study showed that two patients out of 26 Xie; funding acquisition, none; resources, Zhanfang Li; and in experimental group with gentamicin added to epidural analgesics supervision, Pengcheng Xie. were found having candida infection 2 months later (14). Gentamicin is toxic with possible harm to the health of patients. In this study, cefazolin was added to epidural analgesics; one patient in group Data availability C terminated the trial because the analgesics were infected with All data generated or analyzed during this study are included in this colibacillus on day 26. The infection was likely associated with published article. the patient’s unsanitary living environment and habits. The surface of the catheter was infected by colibacillus, and the analgesic was contaminated when the medicine case was changed. The patient did Funding not have symptoms of infection, which was possibly related to the None. bacterial filter connected to the epidural catheter. It could effectively filter bacteria and prevent infection by blocking the microorganism into the epidural cavity. Four patients in group C got infected at Acknowledgements puncture site on day 11, 19, 23 and 27, respectively. The infection None. disappeared after local skin disinfection. No infection was found in group G. Group G had higher overall satisfaction of patients. The addition of cefazolin to analgesics in group G could effectively inhibit bacterial infection and better ensuring the safety of patients. References Due to the small sample size, the exact cause of analgesic infection 1. Vissers KC, Besse K, Wagemans M, et al. Pain in patients with cancer. Pain and the optimal dosage for cefazolin have not been determined. This Pract 2011;11:453–75. needs further research. 2. Su J, Soliz JM, Popat KU, Gebhardt R. Epidural analgesia and subcuta- neous heparin 3 times daily in cancer patients with acute postoperative pain. Anesth Analg 2018;127:e57–9. Conclusion 3. Wiesmann T, Hoff L, Prien L, et al. Programmed intermittent epidural bolus versus continuous epidural infusion for postoperative analgesia after Epidural analgesics combined with cefazolin can be safely applied major abdominal and gynecological cancer surgery: a randomized, triple- to patients with terminal cancers. It has definite analgesic effect, blinded clinical trial. BMC Anesthesiol 2018;18:154. 872 Epidural analgesia for terminal cancer patients

4. Mohamad MF, Mohammad MA, Hetta DF, Ahmed EH, Obiedallah AA, 9. Aloia TA, Kim BJ, Segraves-Chun YS, et al. A randomized controlled trial Elzohry AAM. Thoracic epidural analgesia reduces myocardial injury in of postoperative thoracic epidural analgesia versus intravenous patient- ischemic patients undergoing major abdominal cancer surgery. JPainRes controlled analgesia after major hepatopancreatobiliary surgery. Ann Surg 2017;10:887–95. 2017;266:545–54. 5. Chang WK, Lee MY, Tai YH, Kuo YM, Tsou MY, Chang KY. Does 10. He QH, Liu QL, Li Z, Li KZ, Xie YG. Impact of epidural analgesia on epidural analgesia improve the cancer outcome in hepatocellular carci- quality of life and pain in terminal cancer patients. Pain Manag Nurs noma after resection surgery? A retrospective analysis. J Chin Med Assoc 2015;16:307–13. 2019;82:295–9. 11. Zimmitti G, Soliz J, Aloia TA, et al. Positive impact of epidural analgesia 6. Amsbaugh AK, Amsbaugh MJ, El-Ghamry MN, Derhake BM. Opti- on oncologic outcomes in patients undergoing resection of colorectal liver mal epidural analgesia for patients diagnosed as having gynecologic metastases. Ann Surg Oncol 2016;23:1003–11. cancer undergoing interstitial brachytherapy. J Clin Anesth 2016;35: 12. Vogelaar FJ, Abegg R, van der Linden JC, et al. Epidural analgesia associ- 509–15. ated with better survival in colon cancer. Int J Color Dis 2015;30:1103–7. 7. Day A, Smith R, Jourdan I, Fawcett W, Scott M, Rockall T. Retro- 13. Kinoshita S, Inoue D, Sakuyama T, et al. A case of stage IV b pancreatic spective analysis of the effect of postoperative analgesia on survival in cancer in which a catheter-related infection caused by epidural/subcu- patients after laparoscopic resection of colorectal cancer. BrJAnaesth taneous reservoir therapy affected palliative home-based care. Gan To

2012;109:185–90. Kagaku Ryoho 2012;39:61–3. Downloaded from https://academic.oup.com/jjco/article/50/8/867/5837331 by guest on 01 October 2021 8. Cummings KC III, Zimmerman NM, Maheshwari K, Cooper GS, Cum- 14. Chen YQ, Yuan L, Chen YQ. Effects of epidural analgesia on the quality mings LC. Epidural compared with non-epidural analgesia and car- of life of patients with terminal cancer. Chin J Gerontol 2014;34:304–6. diopulmonary complications after colectomy: a retrospective cohort study 15. Jeon YS, Lee JA, Choi JW, et al. Efficacy of epidural analgesia in patients of 20,880 patients using a national quality database. J Clin Anesth with cancer pain: a retrospective observational study. Yonsei Med J 2018;47:12–8. 2012;53:649–53.