The effect of the local cold application on low back pain and vascular complications of patients undergoing coronary angiography

Fariba Ebrahimi-Shalmani1, Farideh Hasavari1, Salman Nikfarjam2, Ehsan Kazemnejad Leili3, Nazila Javadi-Pashaki*4,5 1Cardiovascular Diseases Research Center, Rudsar Shahid Ansari Hospital, Guilan University of Medical Sciences, Rasht, Iran 2Department of Cardiology, Cardiovascular Diseases Research Center, Heshmat Hospital, School of Medicine, Guilan University of Medical Sciences, Rasht, Iran 3Department of Biostatistics, School of Nursing and Midwifery, Guilan University of Medical Sciences, Rasht, Iran 4Department of Nursing, Social Determinants of Health Research Center (SDHRC), Guilan University of Medical Sciences, Rasht, Iran 5Department of Nursing, Cardiovascular Diseases Research Center, School of Nursing and Midwifery, Guilan University of Medical Sciences, Rasht, Iran

A Coronary angiography due to the risk of vascular complications and low back pain development can lead to undesirable outcomes. This clinical trial was performed on 110 patients undergoing CAG through the femoral artery in 2019. Low back pain was assessed by VAS at baseline, 2, 4, 6 and 24 hours after angiography and vascular complications were investigated 9 times in this period. Results of group and time interaction analysis revealed that pain and vascular complications were signi icantly different between these two groups (P <0.001). After controlling for individual, clinical and technical variables, the scores of low back pain were associated with the intervention group compared to the control group (P <0.001), Body Mass Index (P=0/035) and INR (P <0.001). The extent of hematoma was associated with group (P = 0.003) and also the extent of ecchymosis was associated with group (P = 0.002), education (P = 0.44) and BMI (P = 0.035). In this study, the local cold was effective in back pain and vascular complications in patients undergoing CAG.

INTRODUCTION
Cardiovascular disease are the leading cause of disability and death worldwide (Joseph et al., 2017). According to the World Health Organization reports, deaths due to heart disease are expected to become the most common cause of death in 2030 (Resolution UGA, 2015). The burden of cardiovascular diseases (CVD) will increase steeply in Iran over , mainly because of the ageing population (Sadeghi et al., 2017).
Due to the high mortality and disability of coronary artery diseases, quick and accurate and rapid diagnosis and treatment are crucial (Neishabory et al., 2010). Currently, coronary angiography (CAG) is an important key diagnostic procedure for assessing the coroner and heart anatomy and physiology, conirmation or ruling out the coronary artery diseases, and gathering information to decide on the need for invasive or non-invasive treatment approaches. In fact, CAG is a routine diagnostic tool worldwide (Libby et al., 2007). According to the American Heart Association, these procedures have been increased by 45 percent from 2008 to 2013 (Sha iee and Ebrahimi, 2016;Hejazi et al., 2013).
Although CAG can play an important role in the diagnosis of coronary artery disease, it has serious implications that need to be considered. The patient's experience during this procedure has an important role in the quality of their life and their willingness in taking follow-ups (Darvishpour et al., 2016). Vascular complications are one of the most common adverse effects of CAG with an estimated prevalence of 0.7 to 28 percent (Woodhead, 2008). Hematoma with 47.5% and hemorrhage with 43.4% are the most common complications (Chair et al., 2003;Youse i et al., 2011;Bakhshi et al., 2014). Prolonged rest in the supine position and the pressure resulting from the sandbag can impair the patient's comfort (Botti et al., 1998;Rezaei-Adaryani et al., 2009). Patients tend to change their position regularly to relieve low back pain and increase perceived comfort (Gianakos et al., 2004). This frequent switching might cause vascular damage. Since pain leads to patient's anxiety, fatigue, impatience, discomfort and also sleep disturbances during hospitalization, it can have adverse effects on treatment and their recovery (Javadi et al., 2015). On the other hand, the ef icacy of using sandbags in reducing vascular complications has been questioned (Chair et al., 2003). Although our literature review demonstrated a positive effect of cold on reducing vascular complications (Bayındır et al., 2017;Shimaa, 2016), comprehensive studies on the effect of the cold application on the prevention of low back pain in patients undergoing angiography are few (Shimaa, 2016). One of the important goals of nursing is to reduce the potential complications after diagnosis and treatment procedures during hospitalization which in turn will improve the quality of care and patients satisfaction (Darvishpour et al., 2016). Therefore, it seems necessary to ind a suitable, effective and safe solution for the patients undergoing CAG. On the other hand, considering the limited number of studies on the effect of cold on low back pain, and emphasize on further studies to select the best care strategy with minimal complications in these patients, this study aimed to determine the effect of the local cold application on the site of femoral sheath inser-tion on low back pain and vascular complications in patients undergoing CAG.

METHODS
This study is a randomized clinical trial performed on 110 patients undergoing elective CAG, referred to a teaching hospital in the north of Iran during March -September 2019. Patients were selected base on inclusion criteria including signing written informed consent, older than 18 years old, performing angiography via the femoral artery, not being sensitive to cold, no active hemorrhage in the femoral area during the procedure and before the catheter withdrawal, not experiencing lower back pain before hospitalization, ability to understand Persian language, no psychiatric disorders, not having any vision and hearing problems, not having a history of treatment with thrombolytic drugs and also hemorrhage disorder for less than 6 months, having a normal coagulation test results and not using sedative drugs during the last 24 hours.
Then, they were randomly allocated in two groups intervention (n = 55) and control (n = 55), by using the random block program on the computer (block size of 4).
In the intervention group, the cold pack was used for hemostasis. In the control group, the routine procedure (sandbag) was used. Sample size was determined based on the results of the study by cureek with parameters d=0.25, P2 = 42%, α = 0.05, β = 0.2, P1 = 27.9% using the following formula (Chair et al., 2004).
α 2 Exclusion criteria were the occurrence of any conditions that may lead to impairment in the care process (Myocardial infarction, changes in consciousness level, vasovagal shock, etc.) and diastolic or systolic blood pressure higher than 110 and 190 mm Hg before removing the sheath.
The data collection tools consisted of 5 sections: demographic, clinical, technical information, also the visual analogue scale for pain assessment (VAS) and vascular complications assessment form. Demographic information included age, gender, education, Body Mass Index (BMI), marital status, smoking history, height and weight. Clinical information included: a history of anticoagulant use (drug name, dose), results of the coagulation studies (latest PT, PTT, INR, platelet count and hematocrit), history of comorbidities (hypertension, dia-betes, hyperlipidemia). The technical information included: angiographic duration, sheath size, location and number of venipunctures for catheterization and duration of hemostasis after sheath removal. Low back pain was assessed by VAS. Forms of assessment of vascular complications included: type of complication, amount of hemorrhage, the extent of hematoma and extent of ecchymosis. The amount of hemorrhage was obtained by weighing blooded gases with a precise scales (Standard Feutre, China), with a sensitivity of 0.01 g and an automatic calibration, based on the difference between the weights of dry gas and blooded one. Any palpable mass at the catheter insertion site that protrudes from the body's normal surface was considered as a hematoma. A standard lexible ruler was used to measure these hematomas. The extent of hematoma was calculated by multiplying the large diameter by the small diameter in centimetres. Any discoloration (blue-violet) caused by hemorrhage under the skin in the catheter entry site was considered as ecchymosis. The extent of ecchymosis was obtained by multiplying the size of the largest and the smallest diameter in centimeters.
Patients in both groups underwent cardiac monitoring and their vital signs were controlled, after transfer to the post-angiography ward. The sheath was removed and the duration of hemostasis was recorded by the researcher using a chronometer. After hemostasis control and dressing installation, in control group sandbags weighing 3-4 kg were placed on the position for 4 hours and in the intervention group, cold packs in dimensions similar to sandbag (18 × 14 cm) with a lexible structure which could gain the body's shape, were installed.
The cold packs were placed on the position 4 times, each time for 20 minutes at 10-minute intervals for a period of 2 hours. Patients in both groups were advised to lie in a supine position for 6 hours and not to move the foot under catheterization. The pack's temperature was in the range of 15-18 • C which could be maintained for 30 minutes. A laser thermometer was also used to monitor the temperature of the pack. Low back pain was assessed by VAS at baseline, 2, 4, 6 and 24 hours after the onset of intervention. Vascular complications including Hemorrhage, hematoma, and ecchymosis were assessed 9 times, before installing dressing, every 15 minutes in the irst hour, every 30 minutes in the second hour and after, every hour up to 3 hours. Waterproof markers were used to determine the area of vascular complication, and a lexible transparent ruler was used to measure its extent.

Ethical considerations
This study was performed after receiving written approval of the Research Ethics Committee of Guilan University of Medical Sciences under IR.GUMS.REC.1397.340 and registered in IRAN Registry of Clinical Trials under IRCT 20180701040297N1. Before starting the sampling, the necessary explanations of the research objectives, methods, rights, and expectations were presented to each subject, if desired, they signed a written informed consent to participate in the study.

Data analysis
The data were analyzed by applying descriptive and inferential statistics using SPSS software (version 22). To determine the normal distribution of variables, the Shapiro-Wilk test was used. Chi-square, T-test, and Mann-Whitney U test was used for intergroup comparisons, Repeat measure ANOVA to compare the mean changes of pain scores in the two groups, Bonferroni test to compare pairs of means between the time points, Repeated measures ANCOVA to assess the correlation between the pain scores and intervention type in order to moderate the impact of the confounding variables. The significance level was considered as P <0.05.

RESULTS
The majority of the subjects in the intervention (61.82%) and control (51.73%) groups were over 60 years of age. The demographic, clinical and technical characteristics of the research subjects are presented in Table 1. Our results showed that the intervention and control groups were matched for age, sex, anticoagulant use, history of diabetes, hyperlipidemia, PT, PTT, platelet count and hematocrit (Table 1). Although, there was not any difference in mean of low back pain between the two groups at the onset of the study, a statistically signi icant difference at other time points was observed (P< 0.001).
The assessments of group and time interactions showed a signi icant difference in low back pain between these two groups (P< 0.001) ( Table 2). Investigations on vascular complications showed that there was a signi icant difference in mean hemorrhage at all time points except for the irst 15 minutes and also 3 hours after the sheath removal (P <0.001). There was a statistically signi icant difference between the two groups at all evaluation times in their hematoma and ecchymosis (P <0.001). The assessments of group and time interactions showed a signi icant difference in hematoma and ecchymosis (P <0.001) ( Table 3). Repeated measures ANCOVA test showed a signi icant effect of the   (Table 4). Furthermore, repeated measures ANCOVA test showed the amount of hemorrhage was not affected by any of these variables. Cold pack had a signi icant effect on the extent of hematoma (P = 0.003) and ecchymosis (P = 0.002). Also, it was found a signi icant effect of education (P = 0.44) and BMI (P = 0.035) on ecchymosis.

DISCUSSION
The indings of this study showed that local cold administration after arterial sheath removal at the catheter site is effective in decreasing low back pain and vascular complications after angiography. In this study, the average low back pain in both groups had an increasing trend for 6 hours after angiography and then declined 24 hours later. The average score of pain was lower in the intervention group compared to the control group at all time points. Our results showed that in the control group (sandbag), low back pain scores increased from 1.5 at the beginning of the intervention to 4.1 at the end which was in line with research conducted by Çürük et al. (2017). Also, changes in Back Pain scores in the intervention group (cold pack) increased from 2.5 at the beginning of the intervention to 3.4 at 4 hours (Çürük et al., 2017). Similarly, in the Fathi et al. study, the maximum score of low back pain was at the third hour after sheath removal, and the minimum was at 24. The sandbag removal caused a decrease in the pain in the third hour (Fathi et al., 2017). The possible explanation for the gradual increase in low back pain score is that sleeping in a ixed position causes pressure on the tissues beneath which, in turn, puts pressure on the vas-culature and causes an interruption on blood low, resulting in ischemia and pain (Botti et al., 1998;Rezaei-Adaryani et al., 2009). The increased low back pain in the control group may be due to the fact that the sandbag puts continuous pressure onto the low back muscles, especially the lumbar muscles, causing an increase in their resistance to gravity, decreased strength, and fatigue. Muscle fatigue and decreased blood low together leads to reduced adenosine triphosphate (ATP) and decreased glycogen supply that can increase lactic acid, which eventually results in low back pain (Chair et al., 2004).
Our results showed that low back pain was associated with the intervention group compared to the control, BMI and also the result of the last INR. Although in a study led by Chair et al. there was a signi icant relationship between the subject's weight and their low back pain (P = 0.006) (Chair et al., 2004). However, it is believed that weight gain leads to biomechanical and in lammatory changes, causing vertebral disc impairments and an increase in the pressure on the disc's surface (Sheng et al., 2017). To our knowledge, there is not any study regarding the association between low back pain and INR or other coagulation tests; therefore, we recommend further investigations to explain this phenomenon.
Our results showed that maximum hemorrhage in the intervention group was occurred in the irst 15 minutes and in the control group, 1.5 hours after dressing installation. The average hemorrhage in the intervention group was lower than the control group at all time points. Hemorrhage in the test group reached zero, 3 hours and in the control group, 5 hours after dressing installation which indicates the effectiveness of the cold pack in preventing hemorrhage. In the Fathi et al. study, the maxi-  mum Hemorrhage was at the baseline and the minimum was at 24 hours after angiography. In their study hemorrhage was increasing in the irst 6 hours which is in line with our results (Fathi et al., 2017). In curick's study, hemorrhage was lower in the local cold (3.8%) than the sandbag group (8%) which was statistically signi icant (P = 0.0143) (Çürük et al., 2017). In our study, the average hematoma in the intervention group increased for the irst 15 minutes and then decreased until the end of the study.
In the control group, there was an increasing trend at all time points. Shimaa's study also showed a signi icant difference (p = 0.001) between these two groups (Shimaa, 2016).
In our study, the average ecchymosis increased up until the second 15 minutes of the study and then gradually decreased in the intervention group. Conversely, in the control group, it decreased for the irst 15 minutes and then increased until the end of the study. There was a statistically signi icant difference in ecchymosis between the two groups. In Kurt's study in line with our study local cold application for 15 minutes after sheath removal, reduced the size of the ecchymosis in the next 4 hours (Kurt and Kaşıkçı, 2019). Fathi et al. observed that in the sandbag group, the extent of ecchymosis decreased from the beginning, for 3 hours and then gradually increased until the sixth hour. This trend was similar to our control group results (Fathi et al., 2017).
In Curuk et al. study, the frequency of ecchymosis in the intervention group (7.6%) was lower than that of the sandbag group (18%) and there was a sig-ni icant difference between these two groups (P = 0.0143) (Çürük et al., 2017). Also in the study of Küçükgüçlü and Okumuş showed that the frequency of ecchymosis statistically signi icantly decreased at the site where cold was applied before 2 min after (2 min with 4 min total) the subcutaneous injection of an anticoagulant drug, and the size of the ecchymoses was smaller in patients who had coldapplied (Küçükgüçlü and Okumuş, 2010).
After controlling for confounding variables (individual, clinical, and technical), the hemorrhage was not affected by any of these variables and the groups. Our inding was in line with the results of Curuk's study aimed to determine the effect of the sandbag and local cold application on post-angioplasty vascular complications (Çürük et al., 2017). The extent of hematoma was only in association with the group (cold pack compared to sandbag), which is in line with the study of Kurt and Kasıkcı in angioplasty patients. King's study showed that the application of cold compress was an effective and tolerable method in the treatment of hematoma developing after coronary angiography (King et al., 2008). Furthermore, the extent of ecchymosis was associated with the group, education level, and BMI. Presumably, Lower education levels may cause a lack of accurate understanding to adhere to the instructions which emphasize on inactivity and also bending the corresponded foot after catheter removal leading to an increase in the risk of hemorrhage and subsequent ecchymosis. On BMI, A study conducted by Williams et al. showed a signi icant increase in the odds of vascular complications associated with increased BMI (AOR 1.1, 95% CI 1.0-1.2). Obese people have deeper femoral arteries which are more dif icult to access. This can increase the probability of damage to these arteries while trying to access (Williams et al., 2018). Therefore, we may conclude that ecchymosis also might be in association with obesity.
In general, complications such as hemorrhage, hematoma, and ecchymosis may occur after angiography because of trauma to the vessel. Differences in the reported vascular complications in different studies can be due to the de inition of complications and factors associated with post-angiographic vascular implications such as sheath size, anticoagulant use, age, sex, comorbidities and the type of intervention to control the hemostasis (Chair et al., 2003;Mohammady et al., 2014). Our results showed a signi icant difference between the intervention and control groups in the vascular complication occurrence. We can conclude that local cold can cause vasoconstriction, decreased blood low, and increased coagulation which, in turn, result in decreased hemorrhage, that can ultimately reduce the hematoma and ecchymosis (Shimaa, 2016). Decreased vascular complications in the intervention group ensure greater safety of using cold packs compared to sandbag.
One of the limitations of this study is the lack of evaluation of vascular complications after the 24 hours period following cold pack application which its evaluation is suggested in future studies. In this study, we made an effort to recruit a ixed physician, however, it was not possible to use a ixed nurse to remove the sheath and to control manual pressure applied to establish hemostasis. It is recommended to use a piece of standard equipment for homeostasis control in future studies to ensure gaining constant pressure on the position. Also, it was not possible to blind the researcher for this study.

CONCLUSION
Our Findings showed that patients who received hemostasis with cold pack had less low back pain and vascular complications compared to patients receiving sandbags. According to the results of this study, it can be stated that applying local cold is effective on low back pain and vascular complications in patients undergoing coronary angiography. Therefore, it can be considered as a safe intervention that provides comfort to patients undergoing angiography. It is recommended that policy-makers and clinical nurses consider this procedure as a practical approach to provide comprehensive care.