diff --git a/cashews/backends/interface.py b/cashews/backends/interface.py
index afa2cd72..d97e5458 100644
--- a/cashews/backends/interface.py
+++ b/cashews/backends/interface.py
@@ -146,7 +146,9 @@ async def is_locked(
     async def unlock(self, key: Key, value: Value) -> bool: ...
 
     @asynccontextmanager
-    async def lock(self, key: Key, expire: float, wait: bool = True) -> AsyncGenerator[None, None]:
+    async def lock(
+        self, key: Key, expire: float, wait: bool = True, check_interval: float = 0
+    ) -> AsyncGenerator[None, None]:
         identifier = str(uuid.uuid4())
         while True:
             lock = await self.set_lock(key, identifier, expire=expire)
@@ -163,7 +165,7 @@ async def lock(self, key: Key, expire: float, wait: bool = True) -> AsyncGenerat
                     return
 
                 if wait:
-                    await asyncio.sleep(0)
+                    await asyncio.sleep(check_interval)
                     continue
                 raise LockedError(f"Key {key} is already locked")
             try:
diff --git a/cashews/decorators/locked.py b/cashews/decorators/locked.py
index 737f8958..7e46202b 100644
--- a/cashews/decorators/locked.py
+++ b/cashews/decorators/locked.py
@@ -21,6 +21,7 @@ def locked(
     ttl: TTL | None = None,
     wait: bool = True,
     prefix: str = "lock",
+    check_interval: float = 0,
 ) -> Callable[[DecoratedFunc], DecoratedFunc]:
     """
     Decorator that can help you to solve Cache stampede problem (https://en.wikipedia.org/wiki/Cache_stampede),
@@ -32,14 +33,15 @@ def locked(
     :param ttl: duration to lock wrapped function call
     :param wait: if true - wait till lock is released
     :param prefix: custom prefix for key, default 'lock'
+    :param check_interval: interval in seconds between lock checks while it is waiting for the lock to be released
     """
     ttl = ttl_to_seconds(ttl)
 
     def _decor(func):
         _key_template = get_cache_key_template(func, key=key, prefix=prefix)
         if inspect.isasyncgenfunction(func):
-            return _asyncgen_lock(func, backend, ttl, _key_template, wait)
-        return _coroutine_lock(func, backend, ttl, _key_template, wait)
+            return _asyncgen_lock(func, backend, ttl, _key_template, wait, check_interval)
+        return _coroutine_lock(func, backend, ttl, _key_template, wait, check_interval)
 
     return _decor
 
@@ -50,12 +52,13 @@ def _coroutine_lock(
     ttl: TTL | None,
     key_template: KeyOrTemplate,
     wait: bool,
+    check_interval: float,
 ) -> DecoratedFunc:
     @wraps(func)
     async def _wrap(*args, **kwargs):
         _ttl = ttl_to_seconds(ttl, *args, **kwargs, with_callable=True)
         _cache_key = get_cache_key(func, key_template, args, kwargs)
-        async with backend.lock(_cache_key, _ttl, wait=wait):
+        async with backend.lock(_cache_key, _ttl, wait=wait, check_interval=check_interval):
             return await func(*args, **kwargs)
 
     return _wrap  # type: ignore[return-value]
@@ -67,12 +70,13 @@ def _asyncgen_lock(
     ttl: TTL | None,
     key_template: KeyOrTemplate,
     wait: bool,
+    check_interval: float,
 ):
     @wraps(func)
     async def _wrap(*args, **kwargs):
         _ttl = ttl_to_seconds(ttl, *args, **kwargs, with_callable=True)
         _cache_key = get_cache_key(func, key_template, args, kwargs)
-        async with backend.lock(_cache_key, _ttl, wait=wait):
+        async with backend.lock(_cache_key, _ttl, wait=wait, check_interval=check_interval):
             async for chunk in func(*args, **kwargs):
                 yield chunk
             return
diff --git a/cashews/wrapper/decorators.py b/cashews/wrapper/decorators.py
index 874b8fa2..dafc83f8 100644
--- a/cashews/wrapper/decorators.py
+++ b/cashews/wrapper/decorators.py
@@ -360,6 +360,7 @@ def locked(
         key: KeyOrTemplate | None = None,
         wait: bool = True,
         prefix: str = "locked",
+        check_interval: float = 0,
     ) -> Callable[[DecoratedFunc], DecoratedFunc]:
         return decorators.locked(
             backend=self,  # type: ignore[arg-type]
@@ -367,6 +368,7 @@ def locked(
             key=key,
             wait=wait,
             prefix=prefix,
+            check_interval=check_interval,
         )
 
     def bloom(
diff --git a/tests/conftest.py b/tests/conftest.py
index 29555abb..d3802bb6 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -88,6 +88,7 @@ async def _backend(request, redis_dsn, backend_factory):
         backend = backend_factory(Memory, check_interval=0.01)
     try:
         await backend.init()
+        await backend.clear()
         yield backend, request.param
     finally:
         await backend.close()
diff --git a/tests/test_lock_decorator.py b/tests/test_lock_decorator.py
index 1a828ca9..c15350b4 100644
--- a/tests/test_lock_decorator.py
+++ b/tests/test_lock_decorator.py
@@ -35,6 +35,28 @@ async def func():
     assert mock.call_count == 20
 
 
+async def test_lock_cache_parallel_check_interval(cache):
+    mock = Mock()
+    mock_middleware = Mock()
+
+    async def middleware(call, cmd, backend, *args, **kwargs):
+        mock_middleware()
+        return await call(*args, **kwargs)
+
+    cache.add_middleware(middleware)
+
+    @cache.locked(key="key", ttl=1, wait=True, check_interval=0.05)
+    async def func_interval():
+        await asyncio.sleep(0.01)
+        mock()
+
+    for _ in range(2):
+        await asyncio.gather(*[func_interval() for _ in range(2)])
+
+    assert mock.call_count == 4
+    assert mock_middleware.call_count == 12
+
+
 async def test_lock_cache_iterator(cache):
     mock = Mock()
     chunks = range(10)
diff --git a/tests/test_wrapper.py b/tests/test_wrapper.py
index c29092bf..26175e02 100644
--- a/tests/test_wrapper.py
+++ b/tests/test_wrapper.py
@@ -209,7 +209,7 @@ async def func():
 async def test_cache_lock(cache: Cache):
     m = Mock()
 
-    @cache(ttl=3, lock=True)
+    @cache(ttl=3, lock=True, protected=False)
     async def my_func(val=1):
         await asyncio.sleep(0)  # for task switching
         m(val)