Treatment of osteoarthritis: Nonsteroidal anti-inflammatory drugs (NSAIDs)

The first widely known NSAID was salicylic acid, first synthesized in 1874; its effectiveness in treating rheumatic fever was soon discovered. In 1875, sodium salicylate was first used to treat rheumatic fever. In the mid-1880s, sodium salicylate was widely used as a drug to treat fevers of various origins (malaria, typhus), rheumatic fever, rheumatoid arthritis and gout. A young chemist, Felix Hoffman, who worked in the Bayer Company laboratory in Germany, added an acetyl group to salicylic acid to improve its organoleptic properties. Thus, more than 100 years ago, Bayer first launched Aspirin on the pharmaceutical market, and to this day, acetylsalicylic acid remains one of the best-selling drugs in the world (more than 45 thousand tons per year).

Indomethacin, which appeared on the pharmaceutical market in 1963, was the product of a long-term search for new anti-inflammatory agents. Soon after indomethacin, such drugs as ibuprofen, naproxen, etc. were created.

More than a century after the synthesis of acetylsalicylic acid and 40 years since the introduction of indomethacin to the pharmaceutical market, the NSAID group remains a subject of interest and much controversy, mainly regarding the mechanisms of action and side effects.

The first publication noting the negative effect of acetylsalicylic acid on the mucous membrane of the digestive tract appeared in 1938. Gastroscopy of patients taking acetylsalicylic acid revealed erosions and chronic peptic ulcers. Other side effects of this drug were described somewhat later. The successful use of acetylsalicylic acid in patients with arthritis contributed to the search for drugs that were not inferior to it in effectiveness, but safer, mainly with respect to the digestive tract. Such drugs as phenylbutazone, indomethacin and fenamates were developed. However, all of them, having antipyretic, analgesic and anti-inflammatory effects similar to acetylsalicylic acid, caused side effects characteristic of it. When different chemical groups of drugs have the same therapeutic properties and are characterized by the same spectrum of side effects, it becomes obvious that their activity is associated with the same biochemical process.

For several decades, pharmacologists and biochemists sought the mechanism of action of NSAIDs. The solution to the problem arose during studies of prostaglandins, a group of biologically active substances released from all tissues except erythrocytes and formed under the action of the enzyme cyclooxygenase (COX) on arachidonic acid mobilized from cell membranes. J. R. Vane and co-authors from The Royal College of Surgeons noted that the release of prostaglandins from sensitized guinea pig lung cells was prevented by acetylsalicylic acid. Using the supernatant of the homogenate of damaged guinea pig lung cells as a source of COX, J. R. Vane and co-authors (1971) found a dose-dependent inhibition of prostaglandin formation under the action of salicylic and acetylsalicylic acids and indomethacin.

Further studies using various NSAIDs found that not only did they inhibit COX, but their activity against COX correlated with anti-inflammatory activity. Inhibition of COX, and thus inhibition of prostaglandin formation, came to be considered a unified mechanism of action for NSAIDs.

Thus, the analgesic and anti-inflammatory action of NSAIDs is due to the inhibition of the activity of COX, the key enzyme in the metabolism of arachidonic acid. The first stage of the inflammatory cascade is the release of polyunsaturated fatty acids (including arachidonic acid) linked by an ester bond to glycerol of phospholipids of cell membranes under the action of phospholipases A ₂ or C. Free arachidonic acid is a substrate for the PGN synthetase complex, which includes the active centers of COX and peroxidase. COX converts arachidonic acid into nrG ₂, which in turn is converted into PGN ₂ under the action of peroxidase. Thus, NSAIDs inhibit the conversion of arachidonic acid into PGS ₂. In addition, arachidonic acid is a substrate for 5- and 12-lipoxygenases, catalyzing its conversion into biologically active leukotrienes and hydroxy-icosatetraenoic acids. PGs have pro-inflammatory properties, they increase the permeability of the vascular wall and the release of bradykinins.

The accumulation of PG correlates with the intensity of inflammation and hyperalgesia. It is known that any peripheral pain is associated with an increase in the sensitivity of specialized neurons - nociceptors, which create a signal that is recognized as pain. PGs are a powerful inducer of pain sensitivity. They are not pain moderators in themselves, they are only capable of increasing the sensitivity of nociceptors to various stimuli. PGs switch normal ("silent") nociceptors into a state in which they are easily excited under the influence of any factor.

Of particular interest is the discovery of two COX isoforms, COX-1 and COX-2, which play different roles in the regulation of PG synthesis. The possibility of the existence of two COX forms was first discussed after JL Masferrer et al. (1990) published the results of a study of the effect of a bacterial polysaccharide on PG synthesis by human monocytes in vitro. The authors showed that dexamethasone blocked the increase in PG synthesis under the action of the polysaccharide, but did not affect its basal level. In addition, the inhibition of PG production by dexamethasone was accompanied by the synthesis of a new COX. The two COX isoforms were discovered by molecular biologists studying the neoplastic transformation of chicken embryonic cells. They found that the structure of the inducible form of COX differs from the constitutive form and is encoded by other genes.

Functional activity of COX-1 and COX-2

Function	COX-1	COX-2
Homeostatic/Physiological	Cytoprotection Platelet activation Kidney function Macrophage differentiation	Reproduction Kidney function Bone tissue remodeling Function of the pancreas Vascular tone Tissue reparation
Pathological	Inflammation	Inflammation Pain Fever Proliferative disorder

COX-1 is a constitutive enzyme that is constantly present in the cells of various organs and regulates the synthesis of PGs that ensure normal functional activity of cells. The level of COX-1 activity remains relatively constant, while the expression of COX-2 increases up to 80 times during inflammation. However, there is evidence that COX-1 can also play a role in inflammation, and COX-2 plays a more complex role in the regulation of physiological and pathological processes in the human body. In recent years, the role of COX-2 in the development of not only inflammation, but also other pathophysiological processes, primarily malignant transformation of cells, has been studied.

Although both COX isoforms have the same molecular weight (71 kDa), only 60% of their amino acids are homologous. They also have different cellular localizations: COX-1 is found primarily in the cytoplasm or endoplasmic reticulum, whereas COX-2 is located perinuclearly and in the endoplasmic reticulum.

COX-2 causes the synthesis of PGs, which cause inflammation, mitogenesis, cell proliferation and destruction. Powerful inducers of COX-2 activity are IL-1, TNF, epidermal and platelet growth factors and others, i.e., precisely those biologically active factors that participate in the development of inflammation.

Recently, data have appeared on the significant role of COX-2 in the development of hyperalgesia. According to generalized data, COX-2 mRNA can be induced in the spinal cord after the development of peripheral inflammation. According to the Institute of Rheumatology of the Russian Academy of Medical Sciences, with peripheral inflammation, the level of PGs in the cerebrospinal fluid increases, which are highly sensitive to COX-2 inhibition. Studies in recent years have demonstrated that COX-2 is a natural (constitutive) enzyme expressed in the spinal cord. Thus, COX-2 induces all areas of pain impulse transmission - local, spinal and central.

Thus, the results of recent studies "erase" the clear distinction between COX-1 and COX-2 as constitutive and inducible, as well as physiological and pathological enzymes. It is obvious that both isoforms can induce inflammation in some tissues, and support normal cell function in others.

According to the latest data, the existence of one more isoform, COX-3, is possible. Studying the effects of COX inhibitors in laboratory rats with experimental pleurisy for 48 hours after the injection of the irritant, the authors found that selective COX-2 inhibitors, as well as non-selective COX inhibitors (for example, indomethacin), exhibit anti-inflammatory activity at the beginning of the inflammatory response, which coincides with the expression of COX-2 protein. However, after 6 hours, selective COX-2 inhibitors ceased to act, while non-selective ones continued to exert an effect. At this time, COX-2 protein expression was not observed. The most surprising fact was that after 48 hours, when the inflammatory process was almost completely resolved, COX-2 expression reappeared. This COX-2 protein did not cause the synthesis of proinflammatory PGE ₂ either in the ex vivo experiment with exogenous arachidonic acid or in vivo. On the contrary, at this time, the production in vivo of anti-inflammatory PGs (PGO ₂ and PGR ₂ ), as well as a representative of the cyclopentenone family (ShsohyD ¹²¹⁴ PP ₂ ) was observed.

Inhibition of the new COX isoform by selective and non-selective COX-2 inhibitors between 24 and 48 hours after the stimulus was administered resulted in inflammation not resolving (as in untreated animals) but persisting. According to DA Willoughby et al. (2000), the phenomenon described represents a third COX isoform, COX-3, which, unlike the first two, causes the formation of anti-inflammatory prostanoids.

NSAIDs have been shown to inhibit the activity of both COX isoforms, but their anti-inflammatory activity is associated with inhibition of COX-2.

After studying the three-dimensional structure of COX-1 and COX-2, it turned out that the isoforms differ from each other mainly in the structure of the binding zone with the substrate - arachidonic acid. The active zone of COX-2 is larger than that of COX-1 and has a secondary internal pocket, which plays an important role, since by providing a pharmacological agent with a "tail" complementary to this pocket, it is possible to obtain a drug whose dimensions are too large for the active zone of COX-1, but whose shape corresponds to the active zone of COX-2.

Most known NSAIDs primarily suppress the activity of COX-1, which explains the occurrence of complications such as gastropathy, renal dysfunction, platelet aggregation, encephalopathy, hepatotoxicity, etc.

NSAID-induced side effects may occur wherever PGs are produced, most often in the digestive system, kidneys, liver, and blood system. In elderly people, some changes (decreased production of hydrochloric acid in the stomach, mobility of the stomach and intestine walls and blood flow in them, mucosal cell mass, decreased renal plasma flow, glomerular filtration, tubular function; decreased total body water volume, decreased albumin levels in blood plasma; decreased cardiac output) contribute to an increased risk of developing NSAID side effects. Simultaneous administration of drugs from several groups (especially glucocorticoids), the presence of concomitant pathology ( diseases of the cardiovascular system, kidneys, liver, bronchial asthma) also increase the risk of developing NSAID toxicity.

Research has shown that gastrointestinal symptoms occur in up to 30% of NSAID users. Among elderly patients taking NSAIDs, the rate of hospitalization for peptic ulcers was four times higher than in the same age group of patients not taking NSAIDs. According to the Arthritis, Rheumatism, and Aging Medical Information System (ARAMIS), serious gastrointestinal complications were observed in 733 out of 1,000 patients with osteoarthritis taking NSAIDs for 1 year. In the United States, 16,500 deaths from NSAIDs are recorded among patients with rheumatoid arthritis and osteoarthritis, which is comparable to the mortality rate from AIDS and significantly exceeds the mortality rate from Hodgkin's lymphoma, cervical cancer, multiple myeloma, or asthma. A meta-analysis of 16 controlled studies found that the relative risk of severe gastrointestinal adverse events (those that lead to hospitalization or death) was 3 times higher in people taking NSAIDs than in people not taking NSAIDs. According to the results of this meta-analysis, risk factors for severe adverse events were age over 60 years, history of gastrointestinal diseases (gastritis, peptic ulcer), concomitant use of GCS; the highest risk of developing gastrointestinal adverse reactions was noted in the first three months of treatment.

Side effects of NSAIDs

Side effects from the digestive tract include functional disorders, esophagitis, esophageal strictures, gastritis, mucosal erosions, ulcers, perforation, gastrointestinal bleeding, and death. In addition to the well-known effects of NSAIDs on the gastric and duodenal mucosa, there is increasing evidence of side effects on the mucosa of both the small and large intestines. NSAID-induced enteropathies have been described, which were accompanied by the formation of strictures of the small and large intestines, ulcers, perforation, and atrophy of the mucosal villi. SE Gabriel et al. (1991) described impaired intestinal wall permeability in patients taking NSAIDs.

According to endoscopic studies, NSAIDs can cause erosions and hemorrhages in the submucosal layer in any part of the digestive tract, but most often in the stomach in the prepyloric section and antrum. In most cases, erosive and ulcerative complications of NSAID therapy are asymptomatic.

Recently, a number of studies have established that the mechanism of formation of NSAID-induced ulcers cannot be explained by COX-1 inhibition alone. Of great importance is the direct damaging effect of NSAIDs on the cells of the gastric mucosa with damage to mitochondria and disruption of oxidative phosphorylation, which in turn disrupts energy processes in the cell. It is possible that the formation of ulcers requires the presence of two factors - COX-1 inhibition and disruption of oxidative phosphorylation. Therefore, flurbiprofen and nabumetone - drugs that do not disrupt oxidative phosphorylation - are probably better tolerated by patients compared to other non-selective NSAIDs.

With continuous use of NSAIDs, the development of side effects depends on the dosage and duration of therapy. Taking NSAIDs for 3 months causes side effects from the digestive tract in 1-2% of patients, over the course of a year - in 2-5%.

Currently, the possible role of Helicobacter pylori in the development of NSAID-induced side effects from the digestive system is discussed. It is known that 95% of patients with peptic ulcer of the duodenum are infected with Helicobacter pylori, while in most cases NSAID-induced side effects develop in the gastric mucosa, where the infection rate is 60-80%. In addition, the mechanism of damage to the mucous membrane of the digestive tract by Helicobacter pylori is not associated with the synthesis of PG. Nevertheless, there is evidence that NSAIDs play a role in the recurrence of ulcers, so patients with a history of peptic ulcer are at risk of developing side effects during NSAID therapy. Currently, it is unknown whether eradication of Helicobacter/ry/ori reduces the risk of developing side effects from the digestive system in patients receiving NSAIDs.

NSAIDs may cause adverse renal effects including acute renal failure/prerenal azotemia, renal vasoconstriction, allergic interstitial nephritis, nephrotic syndrome, hyperkalemic/hyporeninemic hypoaldosteronism, sodium and water retention, diuretic resistance, and hyponatremia. However, epidemiological data suggest a low risk of renal dysfunction with NSAIDs.

Risk factors for the development of adverse renal effects in patients taking NSAIDs.

• Presence of kidney pathology
• Diabetes mellitus
• Arterial hypertension
• Congestive heart failure
• Cirrhosis
• Decreased circulating blood volume (taking diuretics, sweating)

The nephrotoxicity of NSAIDs is realized by two mechanisms - inhibition of PG synthesis and idiosyncrasy to NSAIDs. Under normal perfusion conditions, the kidneys do not produce PG, so there are no side effects when using NSAIDs. A decrease in renal perfusion (in chronic renal failure and CHF, dehydration, liver diseases, in old age) is accompanied by the production of PGE ₂ and PP ₂. These PGs induce local vasodilation to maintain normal glomerular blood flow, and also stimulate diuresis, natriuresis and renin release. If such a patient takes NSAIDs, his renal blood flow and glomerular filtration decrease, the secretion of antidiuretic hormone increases, sodium chloride and water are retained, and renin release is suppressed. A state of hyporeninemic hypoaldosteronism occurs, and acute renal failure may develop. NSAID inhibition of COX may also lead to hyperkalemia, especially in patients with concomitant diseases, primarily diabetes mellitus, and to the leveling of the effects of diuretic and antihypertensive therapy.

Allergic interstitial nephritis is a manifestation of idiosyncrasy to NSAIDs, accompanied by fever, skin rash and eosinophilia, occurs 1-2 weeks after the start of NSAID therapy and undergoes regression upon their withdrawal. Other manifestations of idiosyncrasy to NSAIDs include lipoid nephrosis and papillary necrosis.

Despite the fact that hepatotoxicity is a rare manifestation of NSAID intolerance, the frequency of this side effect varies when using different drugs in this group. Thus, liver damage when taking acetylsalicylic acid depends on the dose of the drug and the disease - in systemic lupus erythematosus and juvenile rheumatoid arthritis, hepatotoxicity develops more often than in other diseases. Hepatopathy caused by taking acetylsalicylic acid is often asymptomatic, rarely leads to the development of chronic liver failure and very rarely - to death.

[ 1 ], [ 2 ], [ 3 ], [ 4 ]

Types of NSAID-induced liver injury

Hepatocellular	Cholestatic	Mixed
Acetylsalicylic acid Diclofenac Ibuprofen	Benoxaprofen Nabumetone	Sulindak Piroxicam Naproxen

In addition, there are data on liver damage caused by nimesulide.

Most patients taking drugs of this class belong to the group of elderly people who require constant prevention of acute cardiovascular events. Based on the analysis of 181,441 case histories, W. A. Ray et al. (2002) concluded that despite the combined blockade of COX-1 and COX-2, non-selective NSAIDs do not have a cardioprotective effect (in contrast to low-dose acetylsalicylic acid), so if necessary, they can be prescribed together with acetylsalicylic acid. Thus, ibuprofen blocks the inhibitory effect of low doses of acetylsalicylic acid on the release of thromboxane and platelet aggregation, and the slower-acting diclofenac has delayed similar effects and is therefore better combined with acetylsalicylic acid. At the same time, it was found that coxibs and paracetamol do not compete with low-dose acetylsalicylic acid in terms of disaggregation function. However, acetylsalicylic acid can worsen the tolerability of NSAIDs, as demonstrated in the CLASS study. Thus, when choosing an NSAID for a patient receiving low-dose acetylsalicylic acid, it is necessary to take into account the nature of their interaction.

NSAIDs that cause liver side effects

Very rarely	Ibuprofen
	Indomethacin
	Naproxen
	Oxaprozin
	Piroxicam
Rarely	Diclofenac
	Phenylbutazone
	Sulindak

In recent years, the problem of interaction between NSAIDs and antihypertensive drugs, as well as the use of NSAIDs in arterial hypertension, has become relevant. It is known that due to the suppression of COX-1, which is necessary to maintain many physiological functions, including renal circulation, NSAIDs can neutralize the effect of many antihypertensive agents, especially ACE inhibitors and beta-adrenergic receptor blockers. In addition, the effect of specific COX-2 inhibitors on the cardiovascular system has not been sufficiently studied. In a randomized comparative study of celecoxib (200 mg/day) and rofecoxib (25 mg/day) in more than 800 patients with osteoarthritis receiving antihypertensive therapy for essential arterial hypertension, Welton et al. (2001) found that systolic BP increased in 17% of patients taking rofecoxib and 11% of those taking celecoxib, and diastolic BP increased in 2.3 and 1.5%, respectively. After 6 weeks of treatment, systolic BP increased by an average of 2.5 mm Hg in patients receiving rofecoxib compared with baseline, and even decreased by 0.5 mm Hg in the celecoxib group. The authors concluded that coxibs and antihypertensive drugs are compatible, but celecoxib was better tolerated - edema syndrome and blood pressure destabilization developed less frequently. Almost half of the patients in both groups received diuretics, ACE inhibitors, calcium antagonists, beta-adrenergic receptor blockers as monotherapy from antihypertensive drugs, the remaining patients in each group (48.5 and 44.9%, respectively - celecoxib and rofecoxib) received combination therapy and more than a third (37.9 and 37.1%) in each group - low-dose acetylsalicylic acid. Thus, the results of this study indicate the compatibility of specific COX-2 inhibitors celecoxib and rofecoxib with various antihypertensive drugs or their combinations, as well as a combination with acetylsalicylic acid in the presence of a risk of thrombosis.

In addition to the PG-mediated action, NSAIDs have other effects not associated with PG and COX. Among them is a direct effect on various processes in cells and cell membranes. Thus, NSAIDs inhibit the activation and chemotaxis of neutrophilic granulocytes, reduce the production of free oxygen radicals in them. Being lipophilic substances, NSAIDs are embedded in the lipid bilayer of cell membranes and, thereby preventing interaction between proteins, inhibit signal transmission. Some NSAIDs in vitro inhibit the entry of phagocytes into the inflammation zone.

Along with inhibition of PG synthesis, there are data on other mechanisms of analgesic activity of NSAIDs. These include: central opioid-like antinociceptive action: blockade of NMDA receptors (increase in synthesis of kynurenic acid), change in conformation of alpha-subunits of G-protein, suppression of afferent pain signals (neurokinins, glutamic acid), increase in the content of 5-hydroxytryptamine. The existence of PG-independent mechanisms is indirectly evidenced by data on the dissociation between the anti-inflammatory (COX-dependent) and analgesic (antinociceptive) effects of NSAIDs.

Classification of NSAIDs

A number of NSAIDs affect the synthesis of proteoglycans by chondrocytes in vitro. JT Dinger and M. Parker (1997) proposed a classification of NSAIDs based on their in vitro action on the synthesis of cartilage matrix components in osteoarthritis:

Inhibitory:

• indomethacin,
• naproxen,
• ibuprofen,
• nimesulide,

Neutral:

• piroxicam,
• nabumetone,

Stimulants:

• tenidap,
• aceclofenac.

However, extrapolation of the results of such studies to the human body is questionable. GJ Carrol et al. (1992) performed monthly aspiration of joint fluid from the knee joints of 20 patients with osteoarthrosis taking piroxicam and found a slight decrease in the concentration of keratan sulfate. Although the obtained results may indicate a decrease in proteoglycan catabolism, as the authors emphasize, other interpretations are possible.

Salicylates inhibit phospholipase C activity in macrophages. Some NSAIDs in vitro inhibit the production of rheumatoid factor, prevent the adhesion of neutrophil granulocytes to endothelial cells, and reduce the expression of L-selectins, thereby inhibiting the migration of granulocytes to the inflammation zone.

Another important biological effect of NSAIDs, not related to PG, is the influence on the metabolism of nitric oxide. Thus, NSAIDs inhibit NF-kB-dependent transcription, which leads to the blocking of inducible NO synthase. The latter, induced by proinflammatory cytokines, produces a large amount of NO, which leads to increased signs of inflammation - hyperemia, increased vascular permeability, etc. Acetylsalicylic acid in therapeutic doses inhibits the expression of inducible NO synthase and the subsequent production of NO.

Thus, depending on the nature of COX blocking, NSAIDs are divided into selective and non-selective COX inhibitors. Selective COX-2 inhibitors have a smaller spectrum of side effects and are better tolerated. The relative selectivity of NSAIDs for each isomer is defined as the COX-2/COX-1 ratio and is calculated from the 1C ₅₀ index of the drug for both isoforms, which expresses the concentration of the drug that inhibits PG synthesis by 50%. A selectivity coefficient below 1 indicates relative selectivity to COX-2, while a coefficient above 1 indicates relative selectivity to COX-1.

Classification of NSAIDs based on their ability to selectively block COX-1 or COX-2 activity

Selective COX-1 inhibitors	COX-1 and COX-2 inhibitors	Selective COX-2 inhibitors	Highly selective COX-2 inhibitors
Acetylsalicylic acid in low doses	Most NSAIDs	Meloxicam Nabumetone Etodolac Nimesulide	Celecoxib Rofecoxib Flosulid

Various experimental models are used to determine the COX selectivity of NSAIDs. It should be noted that direct comparison of the results of NSAID selectivity studies obtained in different laboratories is impossible, since the IC ₅₀ values and the COX-2/COX-1 ratio vary greatly even when using the same technique. Such variability may depend on the type of cells used as a model, the type of enzyme preparation, the time of incubation with NSAIDs, the method of COX-2 induction, the protein content in the nutrient medium, etc. For example, nabumetone exhibits COX-2-selective properties in a model using the mouse enzyme in microsomal membranes, but its COX-2 selectivity is insufficient to demonstrate it in models of the human enzyme in cellular or microsomal membranes or in human blood cells ex vivo (Patrignani P. et al., 1994).

Thus, to more accurately assess the selectivity of NSAIDs, it is necessary that the results be confirmed in several models. Studies using human blood cells have proven to be the most indicative. Although the absolute value may vary, the order of the COX-2/COX-1 ratio is generally the same when compounds are tested by several methods.

Non-selective COX inhibitors have not lost their relevance due to their high anti-inflammatory activity and pronounced analgesic effect, but their use is associated with a higher probability of developing side effects.

There are several dozen NSAIDs that are similar in chemical and pharmacological properties and mechanism of action.

To date, there is no clear evidence of the superiority of one NSAID over another in terms of effectiveness. Even if a multicenter study reveals the advantages of a drug in this group, this is often not confirmed in routine clinical practice. However, it is possible to evaluate and compare the tolerability of NSAIDs. Safety is the main feature by which drugs in this group are distinguished.

The multicenter study The LINK Study demonstrated that with long-term use of indomethacin, articular cartilage loss increases 2-fold compared to placebo. Hepatotoxicity is more often observed with diclofenac. Aseptic meningitis is a rare but severe adverse reaction to ibuprofen and sulindac. Cystitis is a complication observed during treatment with tiaprofenic acid; alveolitis can be induced by naproxen, indomethacin causes drowsiness. Changes in blood count, as well as various skin rashes, can occasionally occur with the use of all NSAIDs. According to N. Bateman (1994), among non-selective NSAIDs, ibuprofen and diclofenac are the safest, and piroxicam and azapropazone are the most toxic. However, D. Henry et al. (1996) determined that the tolerability of ibuprofen in high doses did not differ from that of naproxen and indomethacin. At the same time, the effectiveness and safety of propionic acid derivatives served as the basis for the release of over-the-counter dosage forms of these drugs (ibuprofen, ketoprofen and naproxen), which are widely used to relieve pain of various etiologies.

[ 5 ], [ 6 ], [ 7 ], [ 8 ], [ 9 ], [ 10 ], [ 11 ], [ 12 ]

Classification of NSAIDs by chemical structure

I. Acid derivatives
Arylcarboxylic acids
A. Salicylic acid derivatives (salicylates)	B. Anthranilic acid derivatives (fenamates)
Acetylsalicylic acid	Flufenamic acid
Diflunisal	Mefenamic acid
Trisalicylate	Meclofenamic acid
Benorilat	Niflumic acid
Sodium salicylate	Tolfenamic acid
Arylalkanoic acids
A. Derivatives of arylacetic acid	B. Derivatives of heteroaryl acetic acid
Diclofenac	Tolmetin
Fenclofenac	Zomepirac
Alclofenac	Kloperac
Fentiazac	Ketorolac
B. Indole/indoleacetic acid derivatives	G. Arylropionic acid derivatives
Indomethacin	Ibuprofen
Sulindak	Flurbiprofen
Etodolac	Ketoprofen
Acemetacin	Naproxen
	Fenoprofen
	Fenbufen
	Suprofen
	Indoprofen
	Tiaprofenic acid
	Pirprofen
Enolic acids
A. Pyrazolone derivatives (pyrazolidinediones)	B. Oxycams
Phenylbutazone	Piroxicam
Oxyphenbutazone	Sudoxicam
Azapropazone	Meloxicam
Feprazon	Feprazon
II. Non-acidic derivatives
Fluorproquazone	Prokvazon
Flumisol	Tiaramid
Tinoridine	Bufeksamak
Colchicine	Epirizole
Nabumetone	Nimesulide
III. Combination drugs
Diclofenac + Misoprostol
Phenylbutazone + dexamethasone, etc.

Since serious gastrointestinal side effects caused by NSAIDs are dose-dependent, COX-nonselective NSAIDs should be prescribed to patients with osteoarthritis to relieve pain in a low, i.e., “analgesic” dose, which can be increased to an “anti-inflammatory” dose if the first dose is ineffective. For patients at risk, COX-nonselective NSAIDs, even in low doses, should be prescribed in combination with gastroprotectors.

In the 6-month placebo-controlled clinical trial MUCOSA (Misoprostol Ulcer Complications Outcomes Safety Assessment), the addition of the synthetic PG analog misoprostol (800 mcg/day) to NSAIDs resulted in a 40% decrease in the incidence of serious gastrointestinal side effects compared to placebo. At the same time, despite the large number of patients examined (about 9,000 thousand), the reduction in the risk of side effects with misoprostol barely reached statistical significance (p=0.049). Moreover, misoprostol administration was associated with other dose-dependent side effects, in particular diarrhea. Misoprostol at a dose of 400 mcg/day was better tolerated than at a dose of 800 mcg/day, but according to fibrogastroscopy data, it caused a lesser gastroprotective effect.

As an alternative to misoprostol, it is reasonable to use H2-receptor antagonists ₍ eg, famotidine) or proton pump inhibitors (eg, omeprazole). Both groups of drugs have demonstrated efficacy in the treatment and prevention of NSAID-induced ulcers in studies using fibrogastroscopy. However, at usual therapeutic doses, H2 _- antagonists were less effective than misoprostol, whereas omeprazole was not inferior to it in the treatment of NSAID-induced ulcers, was better tolerated, and had a lower recurrence rate.

Meloxicam is a selective COX-2 inhibitor. The safety of meloxicam in vivo and its efficacy in patients with osteoarthritis have been reported in numerous publications.

The main objective of the multicenter, prospective, double-blind, randomized MEloxicam Large-scale International Study Safety Assessment (MELISSA) study was to study the tolerability of meloxicam (the drug Movalis, manufactured by Boehringer Ingelheim, is registered and used in Ukraine) in large, relatively non-randomized groups of patients and to supplement the data obtained in other studies under more limited conditions (Hawkey C. et al., 1998). Diclofenac, a drug with a relatively low level of toxicity to the gastrointestinal tract, was chosen as the comparison drug. Based on the results of studies by M. Distel et al. (1996) and J. Hosie et al. (1996), a meloxicam dose of 7.5 mg/day was recommended for use in a short course during exacerbation of osteoarthritis symptoms. The study included 10,051 patients with osteoarthritis, who were divided into three groups depending on the treatment received (meloxicam - 7.5 mg / day, modified-release diclofenac dosage form - 100 mg / day, or placebo for 28 days). In the group of patients receiving meloxicam, significantly fewer side effects from the digestive system were recorded than in patients treated with diclofenac (Fig. 99). Serious side effects (ulcerogenic effect, ulcer perforation, gastrointestinal bleeding) were observed in 5 patients in the meloxicam group and in 7 patients in the diclofenac group (p> 0.05). Endoscopically, ulcer complications were found in 4 patients receiving diclofenac, while none were found in the meloxicam group. In the meloxicam group, the total duration of hospitalization due to the development of side effects was 5 days, while in the diclofenac group it was 121 days. Among those who refused treatment due to this, 254 patients (5.48%) took meloxicam and 373 patients (7.96%) took diclofenac (p<0.001). Side effects from the gastrointestinal tract were the reason for patients refusing to continue treatment in 3.02% of cases in the meloxicam group and in 6.14% of cases in the diclofenac group (p<0.001). However, a significantly greater number of patients receiving meloxicam refused further treatment due to its insufficient effectiveness (80 of 4635 in the meloxicam group and 49 of 4688 in the diclofenac group, p<0.01). In the group of patients taking diclofenac, a more pronounced positive dynamics in the VAS pain score was also noted than in the meloxicam group. Thus, the results of the study indicate that the tolerability profile of meloxicam is significantly better compared to other NSAIDs, including diclofenac, which may be due to COX-2 selectivity, as well as other reasons (e.g., dose).

A meta-analysis of the results of 10 randomized comparative studies of the efficacy and/or tolerability of meloxicam at doses of 7.5 mg/day and 15 mg/day and reference NSAIDs (piroxicam - 20 mg/day, diclofenac - 100 mg/day, naproxen - 750 mg/day) showed that the former caused significantly fewer side effects compared to the reference NSAIDs (relative ratio - OR - 0.64, 95% CI 0.59-0.69) (Schoenfeld P., 1999). In particular, patients taking meloxicam were less likely to experience ulcerogenic effects, ulcer perforation, and gastrointestinal bleeding (OR=0.52, 95% CI 0.28-0.96), they were less likely to refuse further treatment due to the development of side effects (OR=0.59, 95% CI 0.52-0.67), and were also less likely to complain of dyspepsia (OR=0.73, 95% CI 0.64-0.84).

Nimesulide is an NSAID that is chemically distinct from other representatives of this class by the absence of acidic properties. Nimesulide is a representative of a relatively new group of sulfonanilide derivatives (Bennett A., 1996). Interestingly, nimesulide was initially characterized as a weak COX inhibitor, which was found in various in vitro studies. It was assumed that the "non-prostaglandin" mechanism is more important for nimesulide. According to JR Vane and R. M. Boning (1996), the selectivity coefficient of nimesulide, determined in vitro using an intact cell system, is 0.1.

The pharmacokinetics of the drug is associated not only with its selectivity for COX-2, but also with the peculiarity of its chemical structure (unlike other NSAIDs, nimesulide has weak acidic properties) and half-life (nimesulide - 1.5-5 hours, piroxicam - about 2 days).

Blocking the enzyme phosphodiesterase IV also causes other positive effects of nimesulide:

• inhibition of free oxygen radical production,
• blocking metalloproteases (stromelysin (proteoglycanase) and collagenase)
• antihistamine effect.

The results of numerous studies indicate the high efficacy and safety of nimesulide in patients with osteoarthrosis. In a double-blind, placebo-controlled study, P. Blardi et al. (1991) studied the efficacy of nimesulide in 40 patients with "osteoarthrosis of various localizations" and found the advantage of nimesulide in reducing the severity of joint pain and morning stiffness. In another study with a similar design, RL Dreiser et al. (1991) found a significant advantage of nimesulide compared to placebo in the treatment of 60 patients with osteoarthrosis of the knee joints for 2 weeks according to the VAS pain and AFI Leken, while the incidence of side effects in the group of patients receiving the drug did not exceed that in the placebo group.

The table summarizes the results of controlled studies comparing the efficacy and safety of nimesulide with reference NSAIDs. The duration of treatment in these studies ranged from 3 weeks to 6 months, nimesulide and comparison drugs were prescribed in therapeutic doses, with the exception of the study conducted by V. Fossaluzza et al. (1989), in which the daily dose of naproxen (500 mg) was clearly insufficient.

Celecoxib is the first representative of the coxibs group - specific COX-2 inhibitors. The drug meets all the criteria of a COX-2-specific NSAID - it inhibits COX-2 in vitro and in vivo, exhibits anti-inflammatory and analgesic activity in humans, the dose of the drug required to suppress PG synthesis in the stomach and disrupt platelet aggregation in vivo is many times higher than the therapeutic dose. To inhibit COX-1 activity, the concentration of celecoxib should be 375 times higher than that required to suppress COX-2 activity.

One of the first large comparative studies of the effectiveness of celecoxib (Celebrex, a drug jointly promoted by Pfizen and Pharmacia Corp., is registered in Ukraine) was a study by L. Simon et al. (1999), in which 1,149 patients with osteoarthritis were divided into several groups: celecoxib at 100, 200, and 400 mg twice daily (240, 235, and 218 patients, respectively), naproxen at 500 mg twice daily (225 patients), and placebo (213 patients). The effectiveness of both drugs was significantly higher than placebo. The incidence of gastrointestinal mucosal ulcers detected by endoscopy in the placebo group was 4%, which was no different from that in patients receiving celecoxib (at a dose of 100 mg twice daily - 6%; at a dose of 200 mg twice daily - 4%; at a dose of 400 mg twice daily - 6%; p> 0.05 in all cases). The incidence of gastrointestinal lesions in patients receiving naproxen was significantly higher - 26% (p< 0.001 compared with placebo and all doses of celecoxib).

CLASS (The Celecoxib Long-term Arthritis Safety Study) is a multicenter (386 centers), controlled, double-blind, randomized study of celecoxib tolerability in 8059 patients with osteoarthritis and rheumatoid arthritis. The study drug was prescribed at a dose of 400 mg 2 or 4 times a day, i.e. at a dose 2 or 4 times higher than that approved by the FDA for patients with rheumatoid arthritis and osteoarthritis, while the comparison drugs were prescribed at therapeutic doses: ibuprofen at a dose of 800 mg 3 times a day and diclofenac at a dose of 75 mg 2 times a day. In addition, for the prevention of acute cardiovascular events, acetylsalicylic acid was allowed at a dose below 325 mg / day. The results of the study indicate that the frequency of side effects from the upper gastrointestinal tract when using celecoxib at a dose 2-4 times higher than the maximum therapeutic dose for 6 months is lower than when taking comparison drugs (ibuprofen and diclofenac) at standard therapeutic doses. In patients taking NSAIDs, symptomatic ulcers of the upper gastrointestinal tract and their complications (perforation, stenosis, bleeding) were observed significantly more often than when treated with celecoxib - in the celecoxib group, the frequency of these side effects was 2.08%, in the comparison drug group - 3.54% (p = 0.02). A more detailed statistical analysis revealed no reliable differences in the frequency of complications of gastric and duodenal ulcers between the studied groups (0.76 and 1.45%, respectively, p = 0.09). According to the authors, this was due to the intake of acetylsalicylic acid by some patients (>20%) - among this category of patients, the frequency of complications of peptic ulcers in the celecoxib and comparison drug groups was 2.01 and 2.12%, respectively (p = 0.92), the frequency of symptomatic ulcers and their complications was 4.7 and 6%, respectively (p = 0.49). At the same time, in patients who did not take acetylsalicylic acid, a statistically significant difference was found in the frequency of complications of peptic ulcers between the Celebrex (0.44%) and NSAID groups (1.27%, p = 0.04), as well as the frequency of symptomatic ulcers and their complications (1.4 and 2.91%, respectively, p = 0.02). However, the frequency of adverse effects from the cardiovascular system in the celecoxib and NSAID groups was the same regardless of the intake of acetylsalicylic acid. Thus, according to the CLASS study, celecoxib at doses above the therapeutic dose is characterized by a lower incidence of symptomatic upper gastrointestinal ulcers compared with NSAIDs at standard doses. Concomitant therapy with low-dose aspirin resulted in worsening of celecoxib tolerability in patients with osteoarthritis and rheumatoid arthritis.

Given that celecoxib does not inhibit platelet COX-1 and, therefore, unlike non-selective NSAIDs, does not affect platelet aggregation, the issue of a possible increase in the incidence of cardiovascular events due to hypercoagulation (myocardial infarction, stroke), previously described in patients taking another specific COX-2 inhibitor, rofecoxib, has been widely discussed recently. However, an analysis of a database including more than 13,000 patients treated with celecoxib and the results of the CLASS study in patients with OA and RA did not reveal an increase in the incidence of these complications.

The aim of another double-blind, placebo-controlled, randomized study was to compare the efficacy and tolerability of celecoxib 200 mg/day and diclofenac 150 mg/day in 600 patients with knee OA. The dynamics of the primary efficacy criteria (VAS and WOMAC) during 6 weeks of therapy with celecoxib and diclofenac was more pronounced than in the placebo group. At the same time, no statistically significant difference in efficacy was found between those receiving Celebrex and diclofenac. Side effects were observed in 51% of patients (in the placebo group - in 50%, in the celecoxib group - in 50%, and in the diclofenac group - in 54% of cases).

The occurrence of peripheral edema, flatulence and myalgia were more frequently observed in the celecoxib and diclofenac groups than in the placebo group: Other adverse effects were equally frequent in patients taking celecoxib and placebo. In patients taking diclofenac, adverse effects from the digestive system were recorded more frequently than in the celecoxib and placebo groups (25, 19 and 18%, respectively), including dyspepsia, diarrhea, abdominal pain, nausea and constipation. In addition, in the diclofenac group, a statistically significant increase in the level of liver transaminases, serum creatinine and a decrease in hemoglobin concentration were observed compared with placebo. Such phenomena were not detected in the celecoxib group. It can be concluded that the efficacy of celecoxib at a dose of 200 mg/day in reducing the symptoms of knee osteoarthritis is equivalent to that of diclofenac at a dose of 150 mg/day, but celecoxib is superior to the latter in terms of safety and tolerability.

The results of recent studies indicating the participation of COX-2 in normal kidney development during embryogenesis and maintenance of electrolyte balance require a more in-depth study of the nephrological and cardiovascular side effects of celecoxib. In addition, data have been obtained on a decrease in the hypotensive effect of angiotensin-converting enzyme (ACE) inhibitors by another specific COX-2 inhibitor, rofecoxib, and a dose-dependent increase in arterial pressure and the development of peripheral edema. Therefore, the data of A. Whelton et al. (2000), who analyzed the results of 50 clinical trials involving more than 13,000 patients, about 5,000 of whom took celecoxib for at least 2 years, are of particular interest.

The most common side effects were peripheral edema (in 2.1%), arterial hypertension (in 0.8%), but their development did not depend on the dose and duration of treatment. In general, the frequency of peripheral edema in patients receiving celecoxib did not differ from that in those receiving placebo and was lower than when taking non-selective NSAIDs. The development of edema did not lead to an increase in body weight or an increase in blood pressure both in the group as a whole and in patients with risk factors for this complication, for example, in individuals receiving diuretic therapy. No negative drug interactions were noted between celecoxib and beta-adrenergic receptor blockers, calcium channel blockers, ACE inhibitors and diuretics. All these data provide convincing evidence that celecoxib not only has a favorable gastrointestinal safety profile, but is also well tolerated by patients at high risk of NSAID-induced renal injury and cardiovascular disease. Thus, the development of nephrological and cardiovascular side effects is not a specific property of COX-2 inhibitors and is likely associated with idiosyncrasy to rofecoxib or its metabolites.

Preliminary analysis showed pharmacoeconomic advantages of celecoxib compared with non-selective NSAIDs in patients at risk of developing NSAID-induced severe gastrointestinal complications, taking into account the costs of their prevention (use of misoprostol or omeprazole). For example, in patients with RA without the risk of developing NSAID gastropathy, the incidence of these complications is 0.4%. If we assume that celecoxib reduces the incidence of this complication by 50%, then the prevention of one complication will be observed in only 1 of every 500 patients. At the same time, in elderly patients with a 5% risk of NSAID-induced complications, treatment with celecoxib can prevent their development in as many as 1 of 40 patients. This served as the basis for including COX-2 inhibitors (and primarily celecoxib) in the standard of OA therapy in the USA (ACR, 2000).

The aim of our study was to optimize the quality of treatment based on the inclusion of the COX-2 inhibitor celecoxib in the complex of drug treatment of OA and to study its impact on the quality of life of patients.

Fifteen patients with OA aged 49-65 years were examined; the average duration of the disease was 5.0+2.3 years. The presence of knee joint damage was a mandatory inclusion criterion. X-ray stage II was diagnosed in 10 patients with OA, and stage III in 5 patients. The washout period for NSAIDs was at least 7 days before the start of the study. Patients with OA received celecoxib at a dose of 200 mg/day for 3 months.

To determine the effectiveness of therapy in patients with osteoarthrosis, the Lequesne index, pain according to the VAS, and the success of treatment according to the patient and doctor were assessed. All patients with osteoarthrosis underwent an ultrasonographic examination of the knee joints before and after the course of therapy using the SONOLINE Omnia (Siemens) device with a 7.5L70 linear sensor (frequency 7.5 MHz) in the "ortho" mode in the longitudinal and transverse planes. During the ultrasound, a layer-by-layer assessment of the condition of the joint capsule and its synovial membrane, as well as synovial fluid, hyaline cartilage, bone epiphyses and periarticular tissues was performed.

Quality of life was assessed using the SF-36 questionnaire.

In patients with OA, against the background of celecoxib therapy, the severity of pain according to VAS decreased by 54%, the Lequesne index - by 51%. Patients rated the effectiveness of treatment with celecoxib as very good and good (9 and 6 people, respectively).

According to the analysis of SF-36 scales, the impact of the disease on the emotional state, physical functions and mental health of patients is expressed insignificantly. A large number of positive responses to treatment were noted.

The tolerability of the treatment was assessed as good and very good by both the doctor and the patients. Nausea was observed in 1 patient, pain in the epigastric region and right hypochondrium was observed in 2 patients, and visual acuity was reduced in 1 patient (no objective changes were detected during the ophthalmologist's examination).

All side effects disappeared on their own and did not require discontinuation or reduction of the drug dosage.

In 85% of patients with osteoarthritis, the proposed treatment regimen allowed for complete pain relief, and previously noted synovitis (according to clinical examination and ultrasound) was not detected in any of the patients.

Under the influence of complex therapy, patients significantly improved most indicators of quality of life, especially daily activity and emotional state.

Another representative of the coxibs group is rofecoxib. A series of clinical studies have established the efficacy of rofecoxib in patients with osteoarthritis (at a dose of 12.5 mg/day and 25 mg/day), rheumatoid arthritis (25 mg/day) and low back pain syndrome (25 mg/day). According to a double-blind, placebo-controlled, randomized comparative study of celecoxib at a dose of 200 mg/day (63 patients with knee osteoarthritis) and rofecoxib at a dose of 25 mg/day (59 patients with knee osteoarthritis), after 6 weeks of treatment, no statistically significant differences in the positive dynamics of the main efficacy criteria were found with celecoxib and rofecoxib (p> 0.55), while the changes in the indicators were significantly higher than in the placebo group (p<0.05). The overall incidence of adverse events was similar in the celecoxib and rofecoxib groups, but the former had significantly fewer gastrointestinal adverse events, indicating that celecoxib was better tolerated than rofecoxib at the doses studied.

[ 13 ], [ 14 ], [ 15 ], [ 16 ], [ 17 ], [ 18 ], [ 19 ], [ 20 ], [ 21 ]