diff --git a/src/varint.rs b/src/varint.rs
index 61032d2..db1dadd 100644
--- a/src/varint.rs
+++ b/src/varint.rs
@@ -84,7 +84,7 @@ macro_rules! impl_varint {
 
             fn decode_var(src: &[u8]) -> Option<(Self, usize)> {
                 let (n, s) = u64::decode_var(src)?;
-                Some((n as Self, s))
+                Some((<Self as std::convert::TryFrom<u64>>::try_from(n).ok()?, s))
             }
 
             fn encode_var(self, dst: &mut [u8]) -> usize {
@@ -100,7 +100,7 @@ macro_rules! impl_varint {
 
             fn decode_var(src: &[u8]) -> Option<(Self, usize)> {
                 let (n, s) = i64::decode_var(src)?;
-                Some((n as Self, s))
+                Some((<Self as std::convert::TryFrom<i64>>::try_from(n).ok()?, s))
             }
 
             fn encode_var(self, dst: &mut [u8]) -> usize {
diff --git a/src/varint_tests.rs b/src/varint_tests.rs
index 7a543a7..1da7a52 100644
--- a/src/varint_tests.rs
+++ b/src/varint_tests.rs
@@ -209,4 +209,14 @@ mod tests {
             std::io::ErrorKind::InvalidData
         );
     }
+
+    #[test]
+    fn test_decode_var_too_big() {
+        assert_eq!(u8::decode_var(&0x1FF_u64.encode_var_vec()), None, "u8");
+        assert_eq!(u16::decode_var(&0x1FFFF_u64.encode_var_vec()), None, "u16");
+        assert_eq!(u32::decode_var(&0x1FFFFFFFF_u64.encode_var_vec()), None, "u32");
+        assert_eq!(i8::decode_var(&0x80_i64.encode_var_vec()), None, "i8");
+        assert_eq!(i16::decode_var(&0x8000_i64.encode_var_vec()), None, "i16");
+        assert_eq!(i32::decode_var(&0x80000000_i64.encode_var_vec()), None, "i32");
+    }
 }