The efficacy of paravertebral block evaluated by pain-related biomarkers and reactive oxygen species (ROS) following surgery for breast cancer

A randomized controlled study

Abstract

Background and Aims: 

Paravertebral block has been found to be a recommended method of analgesia for breast surgeries. We aimed to assess the efficacy of paravertebral block compared to morphine by means of the visual analogue scale (VAS), pain-related biomarkers, and reactive oxygen species (ROS) in adult patients undergoing breast cancer surgeries under general anesthesia.

Methodology: 

Forty patients undergoing breast surgery with or without lymph outcome were randomly divided into two groups. Group M received general anesthesia with intraoperative injection morphine (0.1 mg/kg) and group P received general anesthesia with paravertebral block with 0.25% bupivacaine (0.3 ml/kg). The primary objective was to compare the postoperative pain using VAS at baseline, 2 hours, 24 hours, and 48 hours. Secondary objectives were to compare the levels of pain-related biomarkers and ROS in both the groups at baseline, 24 hours, and 48 hours and to study the correlation between the VAS and level of pain-related biomarkers and ROS.

Results: 

There was no significant difference in the postoperative VAS scores in both the groups (P = 0.252, 0.548, 0.488 at baseline, 24 hours, and 48 hours, respectively) and no significant difference in the mean biomarkers in both the groups. The mean VAS negatively correlated with biomarker levels at 24 hours and 48 hours, but the change in VAS after 24 hours and 48 hours did not significantly correlate with the change in the biomarkers.

Conclusion: 

Paravertebral block is equally efficacious to intraoperative morphine for breast surgeries for postoperative analgesia. The biomarker levels were not significantly different when patients received paravertebral block or morphine.

Introduction

Paravertebral block (PVB) is said to be “a well-established option to provide anaesthesia and postoperative analgesia for breast surgery.”[¹] PVB has also been stated to be associated with improvements in long-term survival after breast cancer surgery by modulating the inflammatory and immune response associated with the surgical trauma, reducing opioid and general anesthetic consumption, and promoting cancer cell death through direct effect of local anesthetics. PVB use is associated with lower levels of inflammation and a better immune response in comparison with general anesthesia and opioid-based analgesia.[²] PVB along with the type of surgery is known to attenuate the neuro-endocrine, metabolic, and cytokine responses that affect the immune system.[³⁴] However no previous randomized controlled study has been done to confirm the above findings.

Diaconu et al.[⁵] noted that the biochemical determinations performed for the measurement of ceruloplasmin, circulating immune complexes (CIC), and lipid peroxides (LPO) in breast cancer showed that these parameters can be useful in monitoring the pain intensity. We planned to utilize the dynamics of these serum pain-related biomarkers as well as reactive oxygen species (ROS) to understand the efficacy of PVB in comparison to intraoperative morphine in modulating inflammation and immune response, pain, and tumor growth. We hypothesized that PVB would attenuate these biomarkers more in comparison to the use of morphine.

Hence we planned a study with a primary objective of comparing the VAS scores of patients undergoing breast surgeries—with or without axillary lymph node dissection—receiving PVB or intraoperative morphine at baseline, 2 hours, 24 hours, and 48 hours. Secondary objectives were to compare the serum biomarkers including ceruloplasmin, CIC, LPO, and ROS in both the groups and estimate the correlation between the change in VAS and corresponding change in biomarker levels after 24 hours and 48 hours.

Methodology

Clinical Trials Registry – India (CTRI) registration of the study was done with registration number CTRI/2020/02/023594. Following ethical committee approval, we recruited female patients of 18 years and above of American Society of Anesthesiologists physical status (ASA-PS) I–II undergoing breast surgery with or without lymph node dissection under general anesthesia. Informed consent from the patient or the patient's relatives was taken. The trial was conducted as per CONSORT guidelines (CONSORT flowchart).

Those with bleeding disorders, contraindications, or allergy to any of the study drugs including local anesthetics, non-steroidal anti-inflammatory drugs (NSAIDs), infection at the paravertebral insertion site, pregnancy or breast feeding, obesity, any hearing or visual loss, any neurological or mental disorder, undergoing redo surgery or having significant comorbid illness influencing the measurement of biomarkers like Wilson's disease or Cushing's disease were excluded from the study population.

Following recruitment, the name, age, sex, weight, body mass index (BMI), local address, and in-patient number were noted. Preoperative hemodynamic parameters, comorbid illness like diabetes, hypertension, cardiac diseases, and history of smoking, and ASA-PS were also recorded. Patients were informed and educated about the VAS and its measurement before surgery. All patients underwent baseline evaluation in the week prior to the procedure including complete hemogram, blood urea, serum creatinine, serology, blood sugar, and liver function tests. Preoperative VAS was recorded if pain existed before surgery. Blood was drawn for serum ceruloplasmin level, LPO, CIC, and ROS one day prior to surgery during the pre-anesthetic examination. Blood was also drawn 24 hours and 48 hours postoperatively for these assays.

Forty patients meeting the inclusion criteria were then randomly divided—using simple random sampling through computer-generated random numbers—into two groups with allocation ratio of 1:1 over a period of about two years from November 2019 to September 2021. The group P or group M to be allotted to the patient was written and placed in an opaque sealed envelope which was opened on the morning of the surgery.

Though the sample collection was planned to take place for over six months, due to the constraints of the coronavirus disease 2019 (COVID-19) pandemic it was approved to be completed and extended for over two years. The anesthetist administering general anesthesia, the anesthesiologist evaluating the postoperative pain scores and collecting blood samples, and the laboratory staff conducting the ELISA tests were also blinded to the group allocation. However the sole anesthesiologist administering the PVB was not blinded to the group allocation and did not participate in collection or analyzing the data.

Group P received PVB and general anesthesia. An ideal paravertebral space was identified between T3 and T6 levels in lateral position with operative side up for the procedure. Local anesthesia was administered at 2.5 cm lateral to midline. An 18-G Touhy epidural needle was introduced to contact the transverse process and then advanced cephalad to enter paravertebral space by loss-of-technique technique. Single-shot PVB was administered using 1.5 mg/kg bupivacaine with the total volume of 0.3 ml/kg (maximum 25 ml) before instituting general anesthesia.[⁶] The block was assessed for loss of sensation in the area of surgery and was stated failed if pinprick sensation was present after 10 minutes.[⁷] The anesthesiologist administering general anesthesia was handed a syringe with 10 ml normal saline to be administered intravenously

Group G received general anesthesia with injection morphine 0.1 mg/kg diluted in a 10 ml syringe intraoperatively (maximum 10 mg) without PVB.

The anesthesiologist administering general anesthesia was unaware of the contents of the syringe in either of the groups. They were advised to administer the drug after induction but before incision.

Standard clinical monitoring including electrocardiogram (ECG), non-invasive blood pressure, pulse, saturation, and end tidal carbon dioxide were performed. Patients were pre-medicated with inj. glycopyrollate 0.004 mg/kg, inj. midazolam 0.05 mg/kg, and inj. fentanyl 2 μg/kg and induced with inj. propofol 2 mg/kg and relaxed with inj. vecuronium bromide 0.1 mg/kg. The patient was then intubated with an appropriate sized endotracheal tube. Standard intraoperative care was given with titrated doses of inhalational drug, analgesics, and muscle relaxants and reversed and extubated as per standard protocol

The following parameters were noted including intraoperative and postoperative hemodynamic characteristics, stage of carcinoma breast (TNM classification), and need of lymph node dissection, duration of surgery, and any additional intraoperative analgesics used.

Postoperatively, patients in both the groups received rescue analgesic in the form of inj. paracetamol 1 g IV and repeated at 8 hourly intervals thereafter. The postoperative VAS score at 2 hours, 24 hours, and 48 hours and serum ceruloplasmin level, LPO, CIC and ROS at 24 hours and 48 hours, patient satisfaction with analgesic technique (Yes/No) and any complications including postoperative nausea and vomiting were recorded.

The blood samples were analyzed in the multi-disciplinary research unit of our hospital by ELISA kits procured from Biocodon Technologies Kansas, USA. The assay range of the serum ceruloplasmin kit was 5–2000 ug/ml, serum LPO kit was 0.5–100 nmol/ml, serum CIC kit was 5–600 ng/ml, and that of the ROS kit was 5–4000 U/L.

Based on the previous studies, the mean and standard deviation of VAS between those who received the PVB and those who did not were taken as 1.8 ± 2.1 and 4.1 ± 2.9, respectively,[⁴] and to estimate the true mean difference in VAS after surgery with 95% confidence interval (CI) and 80% power, we required 20 subjects in each group; hence a total of 40 subjects were included in the study.

Continuous variables were summarized as mean ± standard deviation (SD). Categorical variables were summarized as proportion and percentage. Continuous paired data were analyzed using paired t-test and continuous independent data were analyzed using unpaired t-test. Categorical variables were analyzed using χ² test and Fischer's exact test. Correlation between VAS and serum biomarkers was analyzed using the Pearson correlation coefficient.

Results

Twenty patients were recruited in each of the groups. Three patients were excluded from group P and one patient was excluded from group M due to varied reasons (CONSORT flowchart). The two groups were similar with respect to demographic characteristics including age, weight, ASA-PS grading, comorbid illness, and carcinoma breast stage [Tables 1 and 2]. Twenty-seven point five percent of a total of 36 patients recruited for the study belong to stage 3B and the mean duration of the surgery was 120.97 ± 41.26 minutes.

Table 1: Comparison of continuous variables between the two intervention groups

Table 2: Comparison of categorical variables between two intervention groups

Twenty-five percent of patients received single-shot PVB at T3-T4 level, 25% at T4-T5 level, and 50% at T5-T6 level. The mean injection volume of local anesthetic in group P was 18.06 ± 4.68 ml.

The two groups were both similar with respect to the need of lymph node dissection, additional intraoperative analgesia, and patient satisfaction [Table 2]. Lymph node dissection was done in 16 patients in group M and 11 patients in group P. There was no significant hemodynamic fluctuations in both the groups and were comparable.

The mean VAS and biochemical marker levels at different time points are depicted in Table 3. The mean pain scores did not significantly differ in both the groups at baseline, 2 hours, 24 hours, and 48 hours [Table 4].

Table 3: The mean VAS and biochemical marker levels at different timepoints

Table 4: Comparison of pain and biochemical markers (absolute values) between two intervention groups

The mean VAS levels did not significantly correlate with the biochemical markers at baseline but had a significant-to-highly-significant negative correlation at 24 hours and 48 hours [Figure 1]. However when the change in VAS after 24 hours and 48 hours was correlated with the change in serum biomarker levels during the corresponding period, it was not found to be significant [Figure 2].

Figure 1: Scatter plot between mean VAS scores and biochemical markers (actual values) at corresponding timepoints VAS: Visual analogue score; CIC: Circulating immune complexes; LPO: Lipid peroxides; ROS: Reactive oxygen species

Figure 2: Scatter plot between VAS scores and biochemical markers (change values) at corresponding follow-up VAS: Visual analogue score; CIC: Circulating immune complexes; LPO: Lipid peroxides; ROS: Reactive oxygen species

Three patients in group M and one patient in group P had postoperative nausea and vomiting (PONV). No other intraoperative and postoperative complications were noted in both the groups.

Discussion

PVB was recommended as the first choice of regional analgesic technique Grade A by PROSPECT guidelines in 2020 for breast surgeries.[⁸] The use of PVB resulted in lower postoperative pain scores, lower opioid consumption compared to general anesthesia, and a lower incidence of PONV compared to other analgesic techniques.[⁴⁸⁹] However very few studies have evaluated it with respect to other benefits like reduction to immunomodulation, reduction in stress response, concomitant reduction in factors associated with tumor cell angiogenesis compared to opioids.[³] Opioids like morphine are known to impair immune function when given intraoperatively and postoperatively.[³⁹] Perioperative management of pain by unknown means and possibly decreased tumor markers leads to better prognosis of cancer patients.[¹⁰] Biomarkers have been known as clinical tools to evaluate therapeutic interventions and for identification of recurrence and progression of disease.[¹¹] Hence we tried to compare the efficacy of PVB by means of VAS as well as biomarkers.

Özyilkan et al.[¹²] proposed that ceruloplasmin may be used as a tumor marker in the follow-up of patients with breast cancer. Ceruloplasmin levels fall in response to treatment, and those with elevated post-mastectomy ceruloplasmin levels had a higher rate of recurrence.[¹³]

Oxidative stress has also been implicated in the pathophysiology of breast cancers. Kasapović et al.[¹⁴] showed that breast carcinoma is related to increase of lipid peroxidation in plasma. Suppression of AO enzymes associated with breast cancer and aging is most likely the reason for the increased levels of reactive oxygen species (ROS).

Diaconu et al.[⁵] noted that monitoring the dynamics of ceruloplasmin, lipid peroxides, and CIC with that of the anesthetic-surgical act of a breast cancer patient becomes useful for an effective chasing and treatment of pain. Though the chosen parameters are from different metabolic pathways, all of them have a common point in the cascade of events associated with inflammation and pain, and also with the events related to the malign transformation.[⁵] Low levels of ceruloplasmin, LPO, and CIC are also associated with better prognosis.[¹²¹⁴¹⁵] Hence monitoring these parameters in a clinical setting to study the efficacy of a procedure like PVB can be a vital link to investigate the other advantages of PVB.

Through the VAS, our study demonstrated that PVB was equally efficacious in managing postoperative pain as injectable morphine. Many previous studies have noted better postoperative analgesia with PVB compared to opioids.[¹⁶] The difference in the results of our study could be due to a more liberal use of intraoperative morphine 0.1 mg/kg compared to previous studies whilst our aim was to bring out the harmful effects of opioids and beneficial effects of PVB.

The postoperative level of biomarkers as hypothesized by authors was not significantly different in the PVB group. This could possibly be due to different unknown factors other than the anesthesia which could have confounded the results; moreover, the results of the study had to be corroborated with the biomarkers we chose. Possibly a larger sample size with a different array of biomarkers could highlight the surrogate benefits of PVB. Perioperative immune response is affected by multiple factors including surgical and anesthetic plans taken during the performance of the procedure.[¹⁷]

However, many previous studies done earlier have noted that PVB has many surrogate benefits.[¹⁸¹⁹²⁰] In a randomized controlled study comparing general anaesthesia (GA) with morphine and GA with PVB, Looney et al.[²⁰] noted that the angiogenesis factors like vascular endothelial growth factor was modestly lower in PVB growth—a result that is consistent with the hypothesis that anesthetic technique may influence breast cancer outcome. In a study comparing patients receiving GA and PVB, it was noted that PVB attenuates cytokine response such as interleukin (IL)-6, IL-10, IL-12 and interferon (IFN)-γ postoperatively.[¹⁷] Similar results were noted by Deegan et al.[¹⁹] using matrix metalloproteinases when they compared patients receiving propofol/paravertebral and sevoflurane/opioid.

Similarly, in a study by Chen et al. evaluating the effect of thoracic PVB on postoperative analgesia and serum level of tumor markers like carcinoembryonic antigen (CEA), CA 199, CA 125, NSE, CYFRA21-1 and SCC, there was no significant difference in both the groups albeit it was done in lung cancer.[¹⁰]

Though the mean VAS and biochemical markers showed significant correlation at 24 hours and 48 hours, it was a negative one, possibly indicating that factors other than pain were involved. Thirty percent of the participants in our study belonged to stage 3B and above. Post hoc we realized that these patients could be on neoadjuvant chemotherapy which had not been analyzed for bias as the information about chemotherapy had not been collected.[²¹]

The change in VAS levels did not significantly correlate with serum biomarker levels after 24 hours and 48 hours in contrast to the study by Diaconu et al.[⁵] This could possibly be due to the different methodology and biochemical tests used in our study.

Though PVB is associated with the pneumothorax, dural puncture and vascular puncture, we did not encounter any such complications. Also, no significant postoperative complications were noted. We were able to perform PVB in 17 patients and failed the attempt in two patients. The success rate of PVB was noted to be 75%–90% by loss-of-resistance (LOR) technique.[²²]

The limitation of the study is that it is not double-blinded and the insertion of epidural needle in both the groups seems unethical and it may be a cause of bias. We conducted PVB using the LOR technique which is a blind procedure. Further studies of PVB by means of ultrasonography or nerve locator are recommended. Also as stated, the use of neoadjuvant chemotherapy in stage 3B and above for down staging can be a source of erratum.

Conclusion

PVB is equally efficacious to intraoperative morphine for postoperative analgesia measured by means of VAS. However it does not decrease tumor biomarker levels of ceruloplasmin, circulating immune complexes, lipid peroxides, and reactive oxygen species significantly compared to morphine.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form the patient (s) has/have given his/her/their consent for his/her/their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

The authors have received financial support from Rajiv Gandhi Institute of Medical Sciences, Bengaluru as a part of Grant-in-aid for RGUHS advanced research projects amounting to 3,00,000. The authors have also received technical support from the Multi-disciplinary Research Unit (MRU), Karnataka Institute of Medical Sciences, Hubballi for analysis of blood samples.

Conflicts of interest

The authors have received financial support from Rajiv Gandhi Institute of Medical Sciences, Bengaluru as a part of Grant-in-aid for RGUHS advanced research projects amounting to 3,00,000. The authors have also received technical support from the Multi-disciplinary Research Unit (MRU), Karnataka Institute of Medical Sciences, Hubballi for analysis of blood samples.

Acknowledgements

The author would like to thank the Multi-Disciplinary Research Unit (MRU), funded by the Department of Health Sciences (DHR), Government of India, New Delhi for providing technical support and laboratory facility. Authors also acknowledge the support of Dr. Ram Kaulgud, Nodal Officer, Dr. Mahantesh K., Scientist and Mr. Veeresh of MRU, KIMS, Hubballi for designing and executing the experiments.